As an API: OpenAPI document · MCP tool list · API reference.

A skill states a capability generically: what an actor needs to know to perform a task. A Tool is one concrete way to exercise it, with an installation, an invocation and typed inputs and outputs. One skill may be satisfied by several tools, and one tool may satisfy several skills.

They are separate graphs and this is the tools one. The relation between them is satisfies, and it runs from a tool to a skill — this tool is one way to do that, never this skill is a tool.

By instance

Each instance that declares a tools graph has its own page.

How they are invoked, and installed

A tool may declare more than one invocation, so these do not sum to the total.

invocation tools
cli 188
inProcess 35
mcp 30
manual 9
installation tools
none 192
cli 26
service 2

Does every satisfies name a skill that exists?

Yes — all 100 skills named across 219 tools resolve to a skill document in this checkout. A satisfies pointing at nothing would be a tool advertising a capability the graph cannot locate.

Every tool

Reached as:219 of 219 shown
tool what it does invoked satisfies i/o qa
agent-memory · JSON-LD
Agent memory assembler
Assemble every declared memory directory’s nodes into each agent’s memory, per vendor: Claude Code’s .claude/agent-memory/<agent>/MEMORY.md (a marked region, injected when the subagent starts) and Antigravity’s workspace skill .agents/skills/<agent>-memory/SKILL.md. Other vendors (Gemini CLI, Copilot, Codex, Cursor) are not generated. cli agent-memory 0 in / 1 out qa-reports
archimate-actors · JSON-LD
Check that an ArchiMate model’s fhir-actor tags point at the IG’s FHIR ActorDefinitions
Actors come first: the IG’s ActorDefinitions are written from the DAK’s personas, and an ArchiMate model only points at them. For one model a layer’s cat-archimate.config.json names, this reads every element carrying the Archi property fhir-actor, whose value is an ActorDefinition’s canonical URL, and the IG’s ActorDefinitions: FSH InstanceOf: ActorDefinition under input/fsh/, and FHIR JSON under fsh-generated/resources/ or input/resources/, each canonical being <sushi-config canonical>/ActorDefinition/<id> unless its url says otherwise. It fails on a value that is not a canonical URL, a tag that names no ActorDefinition, an element kind that cannot be an actor (a data object, a goal), and a type mismatch (a #person actor on an application or technology element, a #system actor on a business actor or role). It warns about each ActorDefinition no element realizes; --strict makes that a failure. --json prints element → actors and actor → elements, for page back-links. It never writes an ActorDefinition or any FSH, and never matches boxes to actors by name. cli archimate-models 5 in / 1 out qa-reports
archimate-blocks · JSON-LD
Draw a document’s tables and figures from its ArchiMate data, and check the model against the document’s relationship rules
For each table a layer’s cat-archimate.config.json lists under tables, rewrites the block’s Markdown table from an archimate-relationship-rules/v1 file (relationship-rules: source, relationship, target, note; relationship-types: picture, verb, meaning), keeping the caption and footnotes. For each figure under figures that names an svg, writes the view as SVG (the same drawing as the view’s page). For a rules table that names a model, reports the rules the model never uses, the relationships no rule covers, the rules ArchiMate’s relationship table refuses, and the relationship types the rules use without a meaning. The rules are in charge: nothing is rewritten from the model. cli archimate-models 3 in / 1 out qa-reports
archimate-check · JSON-LD
Check a layer’s ArchiMate models against its config
The ArchiMate gate (graph typology archimate): every model a layer’s cat-archimate.config.json names is held, parses (Archi’s native XML or its zipped archive), and resolves — every box in every view draws an element the model holds, every relationship’s ends are in the model — and no .archimate file is held that the config does not name. cli archimate-models 1 in / 1 out qa-reports
archimate-crossings · JSON-LD
Count the tangles in ArchiMate views: lines crossing lines, and lines through boxes
For each view, as render-view.ts draws it, count the pairs of lines that cross and the lines that run through a box they do not connect (groups, and boxes that hold one of the line’s ends, excepted). With --view it lists each tangle by the boxes it joins; with --max it is a gate. cli archimate-models 4 in / 1 out qa-reports
archimate-detangle · JSON-LD
Lower the tangles in ArchiMate views: re-route lines and swap same-size boxes in a row
Searches for a less tangled drawing of each view: a line may be re-routed straight or orthogonally (an L, or a Z through a free lane between boxes), and two same-size boxes in the same parent and row or column may swap places. A change is kept only if it lowers crossings + lines through boxes + lines drawn on top of each other. Without --apply it reports before and after per view; with it, the model is written back in place, only the moved boxes’ bounds and the re-routed lines’ bendpoints changing. cli archimate-models 5 in / 1 out qa-reports
archimate-from-glossary · JSON-LD
ArchiMate model from the glossary
Make an ArchiMate model agree with the instance’s folio-glossary/v1 terms whose focus is an ArchiMate element: elements matched by their skosConcept property (never by name) take the term’s prefLabel and definition, a term with no element gets one on no view, and the relationships the terms list are added. A type that disagrees is reported, never changed. A relationship between two linked elements that no term lists is reported, and removed with --prune along with the lines that draw it. No box or line is moved. Without --write it only reports; --check fails when anything would change. --seed goes the other way once, and only from an ingested model held under library/ (--from): an authored model is never a source of the glossary. cli graph-rendering 7 in / 1 out qa-reports
archimate-metamodel · JSON-LD
Derive the metamodel an ArchiMate model uses, and check it against ArchiMate’s relationship table
For each model a layer’s cat-archimate.config.json names: the element types it uses, with counts; every (source type, relationship, target type) pattern its relationships follow, with counts and ids; whether ArchiMate’s relationship table (Appendix B of ArchiMate 3.2, as Archi holds it, committed as archimate/schemas/archimate-relationships.json in the tools layer) allows each pattern; and the placeholder elements, named after their own type, that a hand-drawn metamodel leaves in a model. Prints a report, JSON, Graphviz DOT or a Mermaid class diagram. archimate-pages draws the same result on each model’s “Metamodel in use” page, so the metamodel is derived, never drawn. cli archimate-models 4 in / 1 out qa-reports
archimate-pages · JSON-LD
Give every ArchiMate view, element and relationship a page and an IRI, and draw every view
Write a JSON-LD node and a thin page for each model a layer’s cat-archimate.config.json names and for every view, element and relationship in it — keyed by Archi’s own ids — plus the normalised model the pages’ one loader draws from, and every view drawn as SVG from the model’s own bounds and bendpoints in ArchiMate’s notation, each box a link to its element. --out writes into a site being built — the data at the graph’s path, the pages at the route of the visualiser the layer declares for this Tool, <locale>/<harness>/<id>/; without it the files go into the graph and --check gates them. cli archimate-models 3 in / 1 out qa-reports
auto-docs-viewer · JSON-LD
Generated index pages
Render index pages for the declared skills, docs and swimlane-glossary directories, one per directory. cli graph-rendering 3 in / 1 out qa-reports
avatars-css · JSON-LD
Avatar stylesheet
Render the declared avatar nodes — an actor’s glyph and colours, including the overlay states — into the stylesheet the site serves. cli site-presentation-assets 1 in / 1 out qa-reports
bean-claim-check · JSON-LD
Check a bean is free, then claim it on every copy
For one bean: fetch every bean copy, read the last three commits on its file, skip it if another agent committed within the window or its last note is a claim, look for an open PR naming it in every listed repository, and only then set it in-progress with a dated claim note as a plain commit on every copy. A rejected push is fetched, rebased and re-checked, never forced. inProcess cli bean-coordination 10 in / 2 out qa-reports
bean-query · JSON-LD
Query the beans knowledge graph
Query the beans knowledge graph using named graph analytics or arbitrary SPARQL 1.1. inProcess mcp bean-coordination
todo-manager
3 in / 2 out qa-reports
beans-cli · JSON-LD
beans CLI
Read and write the work plan with the beans binary. The normal mechanism when it is installed. cli bean-coordination
continual-progress
idle-backlog
pending-show
session-intent
todo-manager
3 in / 1 out qa-reports
beans-manual · JSON-LD
beans, by hand
Read and write the same work plan without the CLI — scripts/beans-fallback.ts, or editing a bean’s front matter directly. Equal standing to the CLI, not a degraded mode. manual cli bean-coordination
continual-progress
idle-backlog
pending-show
session-intent
todo-manager
3 in / 1 out qa-reports
bibliography-page · JSON-LD
Bibliography page
Render every reference in the checkout (each declared references directory’s folio-reference/v1 nodes and each catalogue directory’s Dublin Core records) as the bibliography page and one page per sub-graph, each with its BibTeX, CSL-JSON, RIS and Dublin Core JSON-LD exports and a per-reference export, at the route of the visualiser core declares for it. cli graph-rendering 3 in / 1 out qa-reports
block-actions · JSON-LD
Block edit and feedback links
Each labelled block’s links back to where it can be changed: [edit] opens the block’s Markdown source in GitHub’s editor on main, and [feedback] opens a new GitHub issue about the block, from the folio’s .github/ISSUE_TEMPLATE/block-feedback.yml when it has one (prefilling only the fields it declares: block, section, source, page, url) or as a plain issue whose body carries the same facts. The document build draws these on every block; this Tool prints them as JSON for one block or all. Governed by the block-actions skill. cli block-actions 5 in / 1 out qa-reports
bpmn-checklist · JSON-LD
BPMN checklist projection
Generate, per diagram, a checklist that walks the BPMN as a state machine: each task names the workflow_complete call that records it and ends with next: by node id; each decision lists its labelled branches (or the DMN that computes it); parallel forks name every branch and joins wait for all; start/end events are entry and exit lines. A rendering, never an authority: it says it is generated, and --check fails when one is stale. cli docs-generation 2 in / 1 out qa-reports
bpmn-engine · JSON-LD
BPMN workflow engine
Run a declared BPMN process as an instance: workflow_start opens one for a subject, workflow_next reports what is enabled, workflow_complete records a step or answers a decision (a DMN-computed one from facts) and REFUSES a step that is not enabled. State is committed under the work plan, so which edge was taken is a record, not a memory. inProcess mcp process-state 2 in / 1 out qa-reports
bpmn-render · JSON-LD
BPMN diagram rendering
Render each process diagram to SVG for the documentation site. The .bpmn file is the source of truth; the picture is generated from it, so a diagram and its image cannot disagree. cli docs-generation 1 in / 1 out qa-reports
build-instance-site · JSON-LD
Build, check and (on request) publish a layer’s own site, or a branches/ preview
From a checkout whose layers arrive by remote mount: mount the pinned closure, install, run the layer’s own gates (--pre), compose the harness chrome for THIS layer, run its source steps, build with Jekyll at the base path read from the git remote (/<repo>, or /<repo>/branches/<branch> for a preview), run the post-build steps (mount-instance-docs, pdf-viewer, set-html-lang, publish-id-lookup, rail-standalone-pages), and check that every relative or base-rooted link in every page resolves to a file under the site, reading each page with an HTML parser. Any failing step stops the run; a link that does not resolve means nothing is published. It never decides to publish: only --publish does, and then a root publish keeps branches/ and _render-log/ on the branch, and a preview replaces only its own branches/<branch>/. cli remote-mount 10 in / 2 out qa-reports
c-value-terms · JSON-LD
Rank candidate glossary terms in prose by C-value
The scoring step of the automatic-term-extraction method: count the word n-grams of one or more Markdown or text files (not across punctuation, front matter or code), and rank them by C-value as Frantzi & Ananiadou 1996 define it, so nested multi-word terms are not swamped by the longer phrases that contain them. Tested against the paper’s own worked example (Table 2). Every row is a candidate; it writes nothing and promotes nothing. cli glossary-terms
ontologist
6 in / 1 out qa-reports
cat-harness-schema · JSON-LD
Layer declaration schema
The zod definition of <name>.json — what a layer may declare about itself — and the published JSON Schema generated from it. cli kg-export 1 in / 1 out qa-reports
catalogue-from-references · JSON-LD
Catalogue a library from its bibliography
Write one folio-dublin-core/v1 record and one folio-catalogue-node/v1 item per library entry that the source ledger joins to a CSL-JSON reference, into the folio’s declared catalogue directory, so gen-library-jsonld reads the title from the catalogue’s dc.title instead of falling to the PDF or the slug. The join is exact: a ledger row naming the entry’s structure.json source file, one reference id, and the same sha256 where both record one. Unjoined entries are reported and left alone. Never touches a node it did not write. cli filing-dublin-core 5 in / 1 out qa-reports
check-dsa-hash · JSON-LD
Verify Data Sharing Agreement hash, citation, and vocabulary terms
Verify a Data Sharing Agreement (DSA) directory or single version: checks that every JSON-LD term in agreement.jsonld and dcat.jsonld is defined in the SHA-256-pinned official W3C ODRL 2.2, W3C DCAT 3 (with SPDX 2.2.2 and ADMS), or DCMI Metadata Terms contexts; checks that the ODRL 2.2 Agreement’s uid is the version’s fixed-version URL and its dcterms:source cites the DCAT 3 dcat:Distribution whose spdx:checksum is checked; verifies that the SHA-256 digest of agreement.md matches spdx:checksumValue; and checks dcat:previousVersion and hash distinctness across versions. cli dsa-checksum 1 in / 1 out qa-reports
check-tools · JSON-LD
Do the Tool nodes agree with their skills?
Check every Tool node’s joins: that each satisfies resolves to a real skill and agrees with that skill’s declared contract, that every io port names a declared type, and that no argv input has a type able to express a shell payload. cli code-node-review 0 in / 1 out qa-reports
ci-agent-review · JSON-LD
Run the agent review
Dispatch the index’s agent-review.yml: have every configured LLM review a commit on main and file one summary issue. Started with the standard GitHub CLI; the run is watched with gh run watch. cli adversarial-code-review 2 in / 1 out qa-reports
ci-bootstrap-pages · JSON-LD
Publish bootstrap-tools’ site
Dispatch bootstrap-tools’ pages.yml: stage the README site and graph with scripts/site.ts and commit them onto gh-pages. Started with the standard GitHub CLI; the run is watched with gh run watch. cli kg-to-portal 2 in / 1 out qa-reports
ci-code-quality-gates · JSON-LD
Run the platform index’s code-quality gates
Dispatch the index’s code-quality-gates.yml: every gate over the whole mounted index, its cone, bundle cache and judge. Started with the standard GitHub CLI; the run is watched with gh run watch. cli platform-gates
qa-reports
2 in / 1 out qa-reports
ci-docs-site · JSON-LD
Build and publish the platform docs site
Dispatch the index’s docs-site.yml: security gate, regenerate, compose each layer’s site, build, export the graph, verify, restore previews, publish to gh-pages, judge the built site. Started with the standard GitHub CLI; the run is watched with gh run watch. cli render-kg-to-cdn 2 in / 1 out qa-reports
ci-feature-staging · JSON-LD
Build and publish a branch preview
Dispatch the index’s feature-staging.yml: build a branch’s site and publish it at branches/<branch>/ on gh-pages. Started with the standard GitHub CLI; the run is watched with gh run watch. cli feature-staging 2 in / 1 out qa-reports
ci-health · JSON-LD
Is CI passing on the default branch?
Report each workflow’s state on the default branch, which a checkout cannot see: a red workflow looks exactly like a green one from in here. One API call over the recent run history, not one request per workflow, so it stays inside the unauthenticated 60/hr limit and stays usable at session start. cli ci-health 3 in / 1 out qa-reports
ci-iris-cdn · JSON-LD
Run the IRIS CDN export
Dispatch iris-cdn.yml: the nightly change ingest or the weekly reconciliation of the IRIS CDN (scripts/iris-cdn/run.ts), then verify the signed release. Started with the standard GitHub CLI; the run is watched with gh run watch. cli iris-cdn-exporter 2 in / 1 out qa-reports
ci-iris-cover-pdfs · JSON-LD
Fetch IRIS cover PDFs
Dispatch iris-cover-pdfs.yml: fetch the PDFs the replica’s home page references (scripts/iris-covers/fetch-referenced-pdfs.py) and render their cover previews (scripts/iris-covers/render-previews.py). Started with the standard GitHub CLI; the run is watched with gh run watch. cli iris-dspace 2 in / 1 out qa-reports
ci-merge-main · JSON-LD
Merge main into opted-in PRs
Dispatch the index’s merge-main.yml: merge main into each opted-in PR branch when it moves, and comment where it cannot. Started with the standard GitHub CLI; the run is watched with gh run watch. cli prepare-merge 2 in / 1 out qa-reports
ci-merge-train · JSON-LD
Run the merge train
Dispatch the index’s merge-train.yml: take the queued PRs through the merge train. Started with the standard GitHub CLI; the run is watched with gh run watch. cli merge-queue 2 in / 1 out qa-reports
ci-post-merge-regen · JSON-LD
Regenerate derived files after a merge
Dispatch post-merge-regen.yml (a layer’s caller of layer-post-merge-regen.yml, or the index’s own): regenerate every derived file from its source on main and open the regeneration PR. Started with the standard GitHub CLI; the run is watched with gh run watch. cli docs-generation 2 in / 1 out qa-reports
ci-pr-checks · JSON-LD
Run a layer’s PR checks
Dispatch the layer’s harness-pr-checks.yml, which calls the tools layer’s reusable layer-pr-checks.yml (mount the pinned layers, run every gate of the layer’s dependency harness, compare head against base) and, on main, layer-main-qa.yml. The workflow is the Tool: its helper scripts (run-gates, compare-gates, main-verdict, …) are its steps. Started with the standard GitHub CLI; the run is watched with gh run watch. cli cache-first-ci 2 in / 1 out qa-reports
ci-pr-tests · JSON-LD
Run the tools layer’s own test suite
Dispatch the tools layer’s pr-tests.yml: the unit and standalone tests, sharded, with the failures compared against main’s. Started with the standard GitHub CLI; the run is watched with gh run watch. cli test-engineer 2 in / 1 out qa-reports
ci-publish-context · JSON-LD
Publish a layer’s JSON-LD context
Dispatch the layer’s publish-context.yml, which calls the tools layer’s reusable layer-publish-context.yml to publish the layer’s graph context at the needs it is pinned to. Started with the standard GitHub CLI; the run is watched with gh run watch. cli kg-export 2 in / 1 out qa-reports
ci-publish-subgraph-index · JSON-LD
Publish a layer’s sub-graph index package
Dispatch the layer’s publish-subgraph-index.yml, which calls layer-publish-subgraph-index.yml to package and publish the layer’s sub-graph index. Started with the standard GitHub CLI; the run is watched with gh run watch. cli npm-kg-distribution 2 in / 1 out qa-reports
ci-repin-train · JSON-LD
Run the re-pin train
Dispatch the index’s repin-train.yml: write the standing-consent layer pins and the lock once, and open or refresh the re-pin train PR. Started with the standard GitHub CLI; the run is watched with gh run watch. cli upstream-version-adoption 2 in / 1 out qa-reports
ci-staging-publish · JSON-LD
Publish staged previews
Dispatch the index’s staging-publish.yml: push the previews staged by feature-staging runs to gh-pages. Started with the standard GitHub CLI; the run is watched with gh run watch. cli ref-stewardship 2 in / 1 out qa-reports
ci-standing-repin · JSON-LD
Run a layer’s standing re-pin
Dispatch a layer’s standing-repin.yml: move the pins that carry a standing consent to their newest main. Started with the standard GitHub CLI; the run is watched with gh run watch. cli remote-mount 2 in / 1 out qa-reports
ci-sweeps · JSON-LD
Run the 15-minute sweeps
Dispatch the index’s sweeps.yml: ask each sweep (staging publish, preview verify, re-pin train verdict, cache report) whether there is work, and start the workflow concerned only when there is. Started with the standard GitHub CLI; the run is watched with gh run watch. cli publish-verification 2 in / 1 out qa-reports
compose-docs · JSON-LD
Compose the site’s documentation tree from its declared docs layers
Write the Jekyll source the docs site is built from: the base docs layer, the repository overlay on top (an overlay’s _config.yml merged, every override reported), and each composed layer under its own name. --staging keeps staging-only visualisations; --changed-files stubs layers a branch does not reach. --shell writes the CHROME only — the layers’ Jekyll machinery and assets, no page, no layer, and the host’s generated includes empty — which an IG repository composes its IG into so its own site wears the main site’s chrome. cli harness-tiles 4 in / 1 out qa-reports
content-context · JSON-LD
Content JSON-LD context
Emit the published JSON-LD @context that both populations share — authored block siblings and ingested library/** nodes reference it by URL — generated from its TypeScript definition rather than hand-kept. cli kg-export 1 in / 1 out qa-reports
content-graph-analysis · JSON-LD
Editorial content-graph analysis
Build the block- and section-level editorial dependency graph of one paper from its .ts manifests and report forward references, cross-chapter coupling, sparse or dense sections and isolated blocks, ranked. Reads uses[]/interprets only — the editorial relation, never the formal one. cli content-graph 6 in / 1 out qa-reports
content-graph-build · JSON-LD
Content graph
Build the content graph under a path and report its edges, separated into the EDITORIAL relation an author maintains and the FORMAL one derived from Lean, never as one number, because a single count lets the formal edges overwrite the editorial signal. cli content-validate 2 in / 1 out qa-reports
content-graph-uml · JSON-LD
A paper’s block graph as UML
Draw a paper’s block graph from buildContentGraph: chapters as packages, editorial edges (uses / interprets) solid and formal Lean edges (type / value) dashed purple, never one derived from the other, and each block filled by its formalization status from proof-objects.json when given. One diagram for the paper and one per chapter, each in portrait and landscape, stamped with its source’s hash. Run from a folio: the platform carries no paper. cli content-graph
graph-rendering
4 in / 1 out qa-reports
context-prefixes · JSON-LD
JSON-LD content-key check
Check every committed JSON-LD document that names the content context, or a layer’s own generated context: every plain key is a declared term (never descending into an @json value), and no file path sits under an @id-coerced term. An empty corpus is a finding, never a clean run. Prefixes are checked by jsonld:contexts --check. cli kg-export 1 in / 1 out qa-reports
corpus-search · JSON-LD
Search the corpus and follow the graph
Seed lexically over the authored folio and every declared library, then expand along graph edges, labelling each hit MATCHED or REACHED and by which edge. Counts are split authored vs ingested, and a root that exists but holds no graph nodes is reported rather than read as an empty answer. inProcess mcp corpus-grep 6 in / 1 out qa-reports
declaration-resolution · JSON-LD
Every path a declaration names resolves — directories, assets, visualisers, needs
Read every layer declaration in the checkout and resolve what it names: directories (and those declared from within) through check:declared-dirs, assets through check:declared-assets, visualisers[].renderedBy through check:visualiser-routes, needs through the instance-graph resolver, and visualisers[].coversKinds against the graph typologies registered IN THIS CHECKOUT — its own and those a mounted layer contributes, read from the checkout under inspection rather than from where the harness module sits. A declaration that does not parse is a finding. One verdict per layer and field, naming the gate that decided it. Exit 0 every reference resolves, 1 one does not, 2 could not determine; a certain finding is reported before anything undetermined. cli kg-affordances
kg-separation
2 in / 1 out qa-reports
diig-phases-svg · JSON-LD
DIIG phases, steps and resources chart
Draw a DIIG “phases, steps and resources” chart as SVG from diig-phases-figure/v1 data on a who-iris library entry (DIIG Fig. 1.1.1, or a handbook’s republished version such as the product catalogue handbook’s Fig. 4), or from a folio’s diig-phases-adaptation/v1 file that adds its own tiles and marks itself “(this document)”. Every label comes from the data; tiles link to their library entry or glossary concept. cli figure-rebuild 2 in / 1 out qa-reports
discuss · JSON-LD
discussion
Put a question to a person or a sibling agent and receive an answer, to determine which harness this repository should become and which repositories are read from and written to. The two facts no file holds. manual discussion 2 in / 1 out qa-reports
document-kinds-viewer · JSON-LD
Document kinds viewer
Render each declared document-kinds directory: every kind’s structure, sections and sources, one page per subject layer. cli graph-rendering 2 in / 1 out qa-reports
drain-queue · JSON-LD
Build a bean drain queue
Read every bean copy, keep the drainable open beans (task, bug, feature in todo or in-progress, not blocked, not owner-held), group them by the owner already working in that area, mark the ones another agent touched within the window as skip, and list everything left out with its reason and every bean whose copies disagree. The newest updated_at wins where copies disagree. inProcess cli bean-coordination
idle-backlog
5 in / 2 out qa-reports
dublin-core-render · JSON-LD
Dublin Core renderings
Render every folio-dublin-core/v1 record a catalogue item names as Dublin Core XML (qualified DC, per DCMI’s XML guidelines) and as JSON-LD bound to DCMI Metadata Terms. Both are written into the instance’s published root (<instanceRoot>/dublin-core/<stem>.dc.xml and .dc.jsonld), so the site mount publishes them beside the item pages. Deterministic. --check fails on a missing, stale or orphaned rendering. Governed by the dublin-core-renderings skill. cli dublin-core-renderings 2 in / 1 out qa-reports
external-schemas-viewer · JSON-LD
External schemas viewer
Render the declared external schemas, and which of this instance’s modules conform to each, as one page. cli graph-rendering 2 in / 1 out qa-reports
feature-staging · JSON-LD
Stage a branch’s preview
Publish a branch’s built site to branches/<slug>/ on the publish branch, so a reviewer compares a rendered before and after rather than a description of one. Stamps the commit SHA, and removes the preview when its pull request closes. cli feature-staging 3 in / 1 out qa-reports
figure-coverage · JSON-LD
Figure coverage and computability
For every figure a source publication declares, report whether the folio built from it has a block and how computable it is: missing, excluded (signed off), picture (a PNG crop), linked (a picture linked to a model view), computable (a diagram, a table or a figure rebuilt by ArchiMate, BPMN, DMN, Mermaid, a class diagram or a generator), or cited (a picture kept on purpose, signed off). Exits non-zero when any declared figure is missing; --require-computable also fails on pictures, linked or not. cli figure-rebuild 4 in / 1 out qa-reports
folio-block-qa-summary · JSON-LD
Folio block QA summary
Summarise a folio’s committed per-block QA verdicts into one block-qa.json a staging preview publishes: each block is failing (a FRESH verdict failed, with the worst severity), stale (a verdict predates the block’s current files), passing, or unaudited. Freshness is the QA sweep’s own rule, including the uses-graph hash for graph-scoped criteria. Runs no checker and writes no verdict. cli review-heatmap 3 in / 1 out qa-reports
folio-block-screenshots · JSON-LD
Folio block screenshots
Picture each changed figure, diagram, table, equation or simulator block on the published main site and on a staging build, and compare the two pictures pixel by pixel in Chromium’s canvas. Writes visual-diff.json (folio-visual-diff/v1: per block, the share of pixels changed beyond anti-aliasing, and the before, after and diff pictures) and visual/*.png, which the review page’s visual renderer shows. A side that cannot be pictured (page or anchor missing) is recorded as missing, never drawn blank. Drives Chromium through Playwright. cli visual-diff 5 in / 1 out qa-reports
folio-changeset · JSON-LD
Folio ChangeSet
What changed in a folio between two git refs, block by block: each block added, removed, or changed — and for a changed block, every aspect that applies (renamed, prose, manifest, moved). Keyed on the block label, which the id-unique / id-stable QA criteria guard, not on file paths. cli diff
staging-review
5 in / 1 out qa-reports
folio-init · JSON-LD
Scaffold a folio
Create a new folio repository that uses this platform — folio/, uploads/, library/, the first manifests, the builder shim, agent files, and the link back to the platform. inProcess mcp cli folio-init
getting-started
repo-conversion
9 in / 1 out qa-reports
folio-review-comment-move · JSON-LD
Move a review comment’s status
A review-process task moves one review-comment/1.0.0 todo’s status (address, send back, resolve, adjudicate, withdraw) through transition(), which refuses any move the named BPMN task may not make. The comment is written to the folio’s todos graph (its declared todo-feedback directory) and, with --commit, committed to the edit-set’s FEATURE branch. Refused on the base branch and on a detached HEAD. cli review-comments 9 in / 1 out qa-reports
folio-review-comments · JSON-LD
Folio review comments
Ingest a pull request’s tagged conversation comments (block: <label> on the first line) into review-comment/1.0.0 todos, and write them as review-comments.json. Idempotent over its previous output, whose statuses it keeps. Re-anchors every comment against the head’s blocks, following renamedFrom, and orphans a comment whose block is gone rather than dropping it. cli review-comments 10 in / 1 out qa-reports
folio-review-coverage · JSON-LD
Folio review coverage
Compute the two facts content-change-review.bpmn’s coverage gate (GW_Covered) reads: uncoveredBlocks, the changed blocks with no reviewer verdict on their CURRENT hash, and openDefects. Reads a preview’s changeset.json, blocks.json and review-comments.json. With --commit, writes the ingested folio-review-verdict/v1 verdicts into the todos graph’s declared review-verdicts directory and commits them to the edit-set’s FEATURE branch; refused on the base branch and a detached HEAD. Prints the facts as JSON on stdout. cli review-comments 7 in / 1 out qa-reports
folio-viewer · JSON-LD
Folio viewer
Render each declared folio directory as a browsable page over its published index. cli graph-rendering 2 in / 1 out qa-reports
gates · JSON-LD
The platform’s quality gates
Run the checks CI runs, derived from the workflow rather than listed here. One Tool for all of them, not one per gate: the list is computed from .github/workflows/code-quality-gates.yml at call time, so it cannot drift from what CI actually enforces. cli continual-progress
platform-gates
2 in / 1 out qa-reports
gates-merged · JSON-LD
Gates on the merged tree
Build this branch merged with the current base in a throwaway worktree and run the full bun run cat gates there — the state a merge will actually produce, which neither the branch’s CI nor the base’s CI evaluates. Exit 0 passes, 1 conflicts or fails, 2 could not determine (never read as clean). The working copy is never touched. cli prepare-merge 1 in / 1 out qa-reports
get-graph-stats · JSON-LD
Content graph counts
Node and edge counts for the content graph by provenance and kind, and whether each root exists. Tells ‘nothing matched’ from ‘that population has not been built’. inProcess mcp corpus-grep 1 in / 1 out qa-reports
get-neighbors · JSON-LD
Walk the content graph from one node
List the nodes adjacent to one node, out (what it depends on) or in (what depends on it), to a given depth and optionally over named edge terms only. An unknown edge term is reported, never silently dropped. inProcess mcp corpus-grep 4 in / 1 out qa-reports
gh-pages · JSON-LD
GitHub Pages (gh-pages)
Push a rendered Knowledge Graph to GitHub Pages at a publication root URL — a staging preview (branches/<slug>/) or the release root, the same steps either way — and report the push: a status (pushed, not pushed, could not determine) and one message carrying the commit merged onto gh-pages and the QA result. Its steps are bootstrap-tools’ render-kg-to-github-pages process, which first provisions the target: an orphan gh-pages branch, then Pages switched on to serve it. manual render-kg-to-cdn 4 in / 2 out qa-reports
github · JSON-LD
GitHub (issues, Actions, pull requests)
GitHub as a Tool: the forge that carries an adjudication as an issue, runs the workflow that records it on Actions, and holds the record as a pull request for a maintainer to merge. One forge among possible others; the process names the skill, never GitHub. manual cli translation-adjudication 1 in / 1 out qa-reports
glossary-build · JSON-LD
Glossary build
Build a paper’s glossary index from its manifests and render the LaTeX. --check reports drift instead of writing, comparing everything except the generated timestamp so a re-run is not mistaken for a change. cli document-intake 2 in / 1 out qa-reports
glossary-from-source · JSON-LD
Glossary candidates from a source
Read a source’s glossary and abbreviations sections (an L1 entry’s sections/, or a folio block) into candidate terms of a folio-glossary/v1 scheme, verbatim, each with source: <file>#L<line>. Acronyms fold into the term they expand as notation. The scheme is merged: terms a person owns are never changed. --check writes nothing and fails when the scheme would change. cli glossary-terms 7 in / 1 out qa-reports
glossary-page · JSON-LD
Glossary page
Render every declared glossary — authored SKOS schemes, the swimlane ledger and the terms extracted from knowledge-graph assets — as the glossary page and one page per asset type, at the route of the visualiser core declares for it. cli graph-rendering 3 in / 1 out qa-reports
glossary-table · JSON-LD
Glossary and acronyms tables
Rewrite the glossary and acronyms tables of a document from its folio-glossary/v1 schemes, as glossary/glossary-tables.json (folio-glossary-tables/v1) lists them: terms with their altLabels, definitions and editorial notes; acronyms (notation and prefLabel); or element types, the placeholder types an ArchiMate metamodel view shows, with each term’s definition and skos:example. The block’s caption and footnotes are left alone. --check writes nothing and fails when a table is stale. cli graph-rendering 2 in / 1 out qa-reports
handover-screen · JSON-LD
Screen a hand-over for injected instructions, field by field
Before a model reads a sub-agent’s report, a delegated prompt, a tool result or a comment from someone who is not the principal, screen it against a declared schema whose top-level fields are control (steers what the receiver does) or data (content it reads). A finding in a control field, or any field the strict schema does not declare, is REFUSED, so a report cannot extend the delegator’s plan. A finding in a data field is QUARANTINED: the original is kept and marked, never stripped. The patterns are a deterministic tripwire (instruction overrides, role and turn spoofs, fence breaks, tool-call syntax, hidden Unicode, exfiltration links, pipe-to-shell), a mitigation and not a guarantee. Exit 0 clean, 1 refused, 3 quarantined, 2 could not determine. cli security 3 in / 1 out qa-reports
harness-state · JSON-LD
Per-harness setup position and health
For every harness in the checkout, run bootstrap’s setup check without performing anything or touching the network, and write one folio-harness-state/v1 file: each step in its state, the first step not done (its position) with what to do next, and health badges that are the harness’s own (setup steps; the layer’s committed health report, unknown without one). A recommendation is said, never acted on. cli initialization-steps 2 in / 3 out qa-reports
ig-folio-site-workflow · JSON-LD
Build and publish an IG repository’s own site with its folio-site.yml workflow
Dispatch the IG repository’s folio-site.yml (copied from the IG layer’s ig-repo-site template): on a runner it composes the harness chrome for THIS layer (compose-docs --shell --instance, gen-navbar-include), composes the IG at the root of that Jekyll shell (stage-ig-sites --compose-at-root), builds it with Jekyll so every IG page wears the theme’s side bar, search box and icon row, publishes the layer’s served directories, rails any page Jekyll did not lay out (rail-standalone-pages --foreign-site --instance), stamps the platform that built it (folio-build.json), and on the default branch deploys the whole site to the gh-pages root. It reports platform drift against sibling sites and re-dispatches the IGs this one depends on when a token allows. cli feature-staging
harness-tiles
2 in / 1 out qa-reports
index-config-migrate · JSON-LD
Convert a checkout to a root index.config.json
Write the root index.config.json (folio-index-config/v1) that says which layers the checkout instantiates, where each comes from (source.local or source.remote) and which owns / (site.landing), from the root *.config.json files and the declaration’s remoteMounts. With --write it moves remoteMounts off the declaration into source.remote entries, writes the generated .gitignore block, and renames a legacy <name>.mount-lock.json to index.lock.json. Without a flag it prints the index it would write. Idempotent: a re-run merges what the declaration gained, and refuses an entry the index already holds differently. A root config naming no layer is reported as unmatched-config and not imported, because what to do with it is a person’s decision. cli index-config
remote-mount
3 in / 1 out qa-reports
ingest-extended · JSON-LD
Ingest, with PDF and image extensions
Ingest a PDF into library/ — embedded outline, page text, OCR for scans, and image extraction — using PyMuPDF, tesseract and pypdf with Pillow. cli library-ingestion 3 in / 1 out qa-reports
ingest-stdlib · JSON-LD
Ingest, standard library only
Ingest an upload into library/ using only the Python standard library — archive listings, CSV and spreadsheet records, technical file metadata, and the content sniff that routes a file to its rung. cli library-ingestion 3 in / 1 out qa-reports
instantiate-harness · JSON-LD
Instantiate a harness over a repository that has none: pin, mount, open the setup-only PR, render the preview
The scripted step of the initiation process (getting-started.bpmn, the repo-conversion skill §4). Reads the harness’s declaration at its pin and every needs, transitively (a need beside the layer that needs it unless –repo-of says otherwise, at its default branch’s tip unless –pin says otherwise, its declaration’s name checked), seeds them all into the repository’s index.config.json with the consent given (remote-mount --seed --whole), sets site.landing to the harness, site.agentRun and mountApprovers, mounts the closure (remote-mount: index.lock.json and the .gitignore mounts block), installs it as root workspaces with the hoisted linker, writes the root bootstrap (.github/cat.ts and .github/mount-from-lock.sh from cat-harness-tools templates/root-bootstrap/) and sets the root’s cat script to bun .github/cat.ts, so bun run cat … mounts the layers itself on a later fresh clone, re-renders the agent surface from the mounted generators, runs the harness’s instantiate:<harness> checkout script when a mounted layer declares it, and commits on a branch (default harness/). Then, unless --local, pushes the branch, provisions gh-pages when absent (`pages-bootstrap --provision`), opens the setup-only PR, runs `prepare:` when declared (what the site is built from and nobody commits, such as an IG's AST) and renders the branch's preview at gh-pages branches// (`stage-local`); with --local the site is built and composed and nothing is pushed. It imports only node:*, so it runs from a fresh clone of cat-harness-tools before anything is mounted: it lays the platform's own closure, at the pins the repository will mount, into a temporary root to run the mount from. On a repository with nothing mounted yet, run the same file from a fresh clone: `git clone --depth 1 https://github.com/litlfred/cat-harness-tools /tmp/cht && bun run /tmp/cht/scripts/instantiate-harness.ts …`. Two more flags, not declared as ports because no constrained type carries them yet: `--pin =` pins one layer of the closure, `--repo-of =<owner/repo>` says where a need lives. cli remote-mount
repo-conversion
8 in / 1 out qa-reports
iris-pages · JSON-LD
who-iris catalogue and docs pages
Render who-iris’s docs pages, the catalogue visualiser at its declared route (what the catalogue knows, and what it records that it does not know), the library’s withheld list, and the IRIS replica’s home page. cli graph-rendering 2 in / 1 out qa-reports
journey-refs · JSON-LD
Every reference a journey or a workshop names resolves
Read every user-scenario and workshop node in the checkout through the layers’ declared directories, with the journeys viewer’s own resolver, and resolve what each names: actors (personas, a step’s lead, experiences[].actor, handoff.from/to, illustrationStyle.cast[].actor), user stories, realisedBy BPMN files (instance-relative) and the element ids in them, uses blocks in a directory declaring building-blocks, req: requirements, FHIR examples and a step’s dak assets (data sets, decision tables, indicators) in a fhir-artifact-index (could not determine while the index is built at publish), illustration.src files, follows/partOf/parts/includes journeys (parts and includes forming no cycle; an includes held against the call activity its step’s realisedBy names, could not determine when the call names no process), the trustLadder code list and each record.state code of it, and ArchiMate views; and each workshop’s scenarios, steps[] {scenario, step} and <scenario>#<step> refs; each device screen’s owner and personCard actors, its steps[] {scenario, step}, and the FHIR examples its qr and recordList blocks name; and every persona’s avatar image path. Every journey is a DAK user scenario (the kind specialises it): its DAK personas, the roles its persona actors play, must be non-empty and each a declared role; a persona may be an actor or (0.4.0) a role. What is derived is not also written: a listed part’s follows or partOf, personas equal to the parts’, a lead equal to the lanes’ first actor, are each a redundant finding, and a 0.4.0 step with no experiences whose lanes no persona plays is a finding. A node that does not parse is a finding. One line per finding, naming the layer, file and step. Exit 0 every reference resolves (or no layer holds a journey or a workshop, which is said); 1 one does not; 2 the declarations could not be read. cli journey-maps 2 in / 1 out qa-reports
journeys-viewer · JSON-LD
Journeys viewer
Draw every journey (user-scenario) and workshop (workshop) in the checkout as static pages at the route of the visualiser a layer declares for this Tool, <locale>/<instance>/<id>/: an index of all journeys, grouped by layer (each a sub-graph page), each parent’s parts nested under it in its order with optional and repeating parts marked (an older file’s partOf children after them, ordered along follows), and a plain DAK user scenario with no steps listed as one; one page per journey, its parts in order and the journeys it is a part of, a journey map with a lane and a feeling curve per persona, then each step, linked to the journey it includes, with its BPMN drawing (the step’s flow nodes highlighted), user stories, requirements, FHIR examples, building blocks, the DAK assets it uses (data sets, decision tables and indicators, linked into the IG) and ArchiMate views; and the workshop layer, the questions for the room on each step, a page per workshop and per module, job aids, and printable A3/A4 sheets; and each device screen (device-screen) that shows a step a workshop presents, drawn as a static device mock (phone, tablet, scanner, desktop, kiosk) with its owner beside it, in the workshop’s Device Demo, linked both ways with the walkthrough. A persona’s picture (meta.persona.avatar) is drawn wherever the persona appears. Layers below the declaring one are read through their declarations, so a downstream site shows the journeys of the layers it mounts too. cli journey-maps 2 in / 1 out qa-reports
judge-tree · JSON-LD
Judge a tree and record the verdict by digest
Run the layer PR-check gates and tests on a composed tree (head) and on its base, or reuse a stored verdict whose digest is the base tree’s, compare them with the layer-pr-checks compare scripts, and write a verdict record: {treeDigest, runner {kind agent|github-actions|local, id}, startedAt, finishedAt, gates [{name, exit, seconds, skipped?}], tests {total, failed, new}, verdict green|red|unknown}. With a token it publishes the verdict as the commit status judge-tree; check answers whether any runner judged a digest green; sample re-runs a seeded few gates of a record. cli platform-gates
prepare-merge
10 in / 1 out qa-reports
kg-audit · JSON-LD
Knowledge-graph audit
Audit every join in the actor→role→skill→task sentence and write a committed QA sidecar per node. The sidecar distinguishes “unbound since it was drawn” from “broken in the commit under review”. cli github action code-node-review 2 in / 2 out qa-reports
kg-graph-export · JSON-LD
Knowledge-graph export
Dump this layer’s knowledge graph — skills, BPMN activities and their lanes, roles, actors, directories — to one JSON-LD document for publication. The export is data; something else draws it. cli bootstrap-graph-emission
bootstrap-graph-publication
kg-export
2 in / 2 out qa-reports
kg-retrieve-npm · JSON-LD
Retrieve a remote Knowledge Graph via npm package or tarball
Retrieve and inspect a remote Knowledge Graph packaged as an npm package or tarball (.tgz), supporting both unhydrated source graph views (authored declarations, skills, schemas) and hydrated materialized graph views (subgraph JSON-LD, metadata indexes, dereferenced graph projections). cli npm-kg-distribution
remote-mount
4 in / 2 out qa-reports
kg-validate · JSON-LD
Validate a node in the graph
Check one file against the schema for its graph typology. ONE tool for every schema: the kind is resolved from the declared directory that holds the file. cli kg-navigation 2 in / 1 out qa-reports
kg-viewer · JSON-LD
Knowledge graph viewer
Render the layer’s published knowledge graph as one page: facets by kind and sub-graph, search, a node’s properties, its edges and back-links, and the graph’s own gaps — reading the JSON-LD document at the site root rather than carrying a copy. cli graph-rendering 2 in / 1 out qa-reports
l1-complete-check · JSON-LD
L1 source completeness
Is a library/<bib-slug>/ entry complete as L1 source content? Each requirement is met, unmet, or NOT-DERIVABLE. cli library-ingestion 1 in / 1 out qa-reports
l1-coverage · JSON-LD
L1 extraction coverage
Of the normative sentences in a publication (the closed marker list of the methodology l1-coverage §1), count how many an L1 extraction CAPTURED, how many are EXCLUDED for a fixed-list reason a person signed off, and how many are UNACCOUNTED — per page and in total. Writes the contract’s §4 report and exits non-zero below 100% accounted-for. cli l1-coverage 4 in / 1 out qa-reports
library-viewer · JSON-LD
Library viewer
Render each declared library directory — its entries, intakes and avatars — as a page per subject layer, and give every entry its own path IRI (///): a materialized thin shell that loads the entry from the published projection, with its JSON-LD manifest as the alternate. cli graph-rendering 3 in / 1 out qa-reports
log-message · JSON-LD
Log a message to the discussion
Write a log line where the human actor will read it: the discussion you are already in. Takes the five required fields and the optional body, and renders them as one entry. manual log-message 6 in / 1 out qa-reports
lsi-index · JSON-LD
Build the Latent Semantic Indexing index of a prose graph
Build the per-graph LSI index sidecar for every declared prose graph (or the one named): input fingerprint, parameters, dimension summaries, nearest neighbours and near-duplicate findings, never the vectors. Records each run’s outcome and input fingerprint, success or failure, as a folio-tool-run/v1 record. cli lsi-indexing 2 in / 1 out qa-reports
lsi-query · JSON-LD
Latent Semantic Indexing over a prose graph
Find units of a declared prose graph (a library, the skills, the beans, docs) that discuss a query in OTHER words — the vocabulary gap lexical search cannot reach. Each hit is labelled lexical+latent or latent only, and the score is a cosine in the latent space, never merged with a lexical result. inProcess mcp cli lsi-indexing 3 in / 1 out qa-reports
lsi-viewer · JSON-LD
Indexing evidence viewer
Render the Indexing evidence page: for each indexing method, the held papers it rests on and the skills, Tools and processes that apply it, and the indexing skills with no evidence yet; and its Keywords page (keywords/), the derived keywords graph as an A–Z index with the six extraction variants side by side and their agreement with headings, author keywords and the glossaries. Index health is the lsi-indexes check of bun run cat health. cli graph-rendering 2 in / 1 out qa-reports
memory-checkpoint · JSON-LD
Keep where the work stands
Before you stop or hand over, keep where the work stands: what is done, what is next, and anything the next person must know to carry on. inProcess mcp work-context 2 in / 1 out qa-reports
memory-note · JSON-LD
Keep a note on the work
Keep a short note on the work you are doing — what you did, what you decided and why, and what is left open — so it is there when the work is picked up again. inProcess mcp work-context 2 in / 1 out qa-reports
memory-recall · JSON-LD
Read what was kept
Read what has been kept about this work so far, oldest first, so you can start where the last person stopped. inProcess mcp work-context 1 in / 1 out qa-reports
merge-guard · JSON-LD
Judge whether a pull request may land
The enforced merge guard: asks GitHub’s live facts about one pull request whether it is finished (its checks, its tier, the session that marked it ready) before it reaches its base, and lands it only when every check passes. With --status it posts the verdict as the merge-guard commit status on the head, which is what the layer’s merge-guard workflow does. cli github action ci-health 5 in / 1 out qa-reports
merge-leftover · JSON-LD
Merge leftover (has a PR’s intent landed?)
After a train merged, compare a PR’s head with the base path by path and say whether what it still changes is ONLY generated files, generated README regions, or changes the base already carries (its patch applies in reverse to the base): landed, not-landed with the authored paths still different, or could-not-determine, which is never shown as clean. Only reports; closing the PR stays a steward action. cli merge-conflict-patterns 2 in / 1 out qa-reports
merge-overlap · JSON-LD
Merge overlap (conflict prediction)
For the open PRs (via gh, or a list of branches), report which pairs would conflict: pairwise overlap on AUTHORED paths, with generated paths excluded using the merge-conflict-patterns declaration; which PRs touch a shared declaration (a layer’s <instance>.json, roles.json, package.json, bun.lock, schemas, BPMN/DMN); and which touch the harness layer’s directory or its tools layer’s. A PR that could not be measured makes no pair independent. JSON (merge-overlap/v1), the conflict-prediction input for composing trains. cli coordinate
merge-conflict-patterns
2 in / 1 out qa-reports
merge-train · JSON-LD
Merge train
Build a train branch from a base SHA: merge each member (a PR number or branch) with merge-base.ts --no-regen, refusing — never hand-resolving — a member whose conflicts no declared pattern covers; then one bun run cat regen, check:l1-complete --write, every check a layer declares afterMerge (its declared writer run when red), and kg:audit:all:check; then merge origin/main, taking main’s side of generated conflicts and regenerating once more. Emits a merge-train-report/v1 JSON report. Never pushes, opens or merges a PR. cli merge-conflict-patterns
prepare-merge
5 in / 1 out qa-reports
mesh-authority · JSON-LD
Validate draft MeSH headings against NLM’s authority
Stage three of skill-pipeline-subject-indexing for MeSH: look up every candidate heading of a draft (a preferred heading, an entry term, or a descriptor id, with an optional qualifier) in NLM’s MeSH authority at id.nlm.nih.gov, current year. Each candidate comes back resolved (in the preferred form), qualifier-not-allowed (descriptor kept, qualifier dropped), ambiguous (options listed for a person) or not-found (dropped). It never adds a heading the draft did not propose and never emits one the authority lacks. cli mesh-subject-indexing 2 in / 1 out qa-reports
methodologies-viewer · JSON-LD
Methodologies viewer
Render the declared methodology graph as one page. cli graph-rendering 3 in / 1 out qa-reports
mount-from-lock · JSON-LD
Replay a committed mount lock with nothing but Node
Lay down every remote mount index.lock.json records — each layer fetched at the lock’s full SHA (sparse, blob-filtered), each directory verified against its treeDigest, a mismatch undone and reported — importing only node:*, so it runs on a fresh clone before any layer is present. It never decides what to mount: remote-mount writes the lock (and checks trust); this only replays it. Refuses to write over a path with tracked files or bytes that no longer hash to the lock. Exit 0 mounted/current, 1 missing, 2 could-not-determine. cli kg-separation
remote-mount
2 in / 1 out qa-reports
narrative-queue · JSON-LD
What narratives are waiting on a person
List the agent-drafted narratives awaiting human confirmation, numbered, with the numbered rejection reasons beside them. The queue is the only place a draft’s state is visible before someone accepts it. cli library-ingestion 0 in / 1 out qa-reports
navbar-include · JSON-LD
Write the site sidebar’s harness navbar include
Render _includes/generated/navbar-footer.html from docs/_data/harness.json with the same renderer every railed page uses. With --instance, render the navbar of an IG repository’s OWN site instead — that layer first, then what it needs; its own pages at this site’s root, every other link to the main site at --link-root — into the shell that site is built from. cli harness-tiles 5 in / 1 out qa-reports
node-kind-pages · JSON-LD
Node kind pages
Render a page for every node kind, every harness holding nodes of it, and every node, under ////. cli graph-rendering 2 in / 1 out qa-reports
node-view-coverage · JSON-LD
Does every node have an IRI, a .jsonld, a page and a row?
For each node a declared directory holds, per layer × folder × kind: whether it has an IRI in its layer’s export (extensionless), its JSON-LD at <iri>.jsonld (and .json where its source is JSON), a view page at <locale>/<harness>/<vis>/<instance>/<id>/, and a row on its folder page that links to it; plus the folder page’s dead links and raw-repository links, and whether a viewer is declared. Folder pages are read with their data files, so rows drawn by JavaScript count. Writes node-view-coverage/v1, one file per layer, under test/results/node-views/. cli github action kg-viewer 4 in / 1 out qa-reports
npm · JSON-LD
Package manager and registry client for npm and GitHub Packages
Package, publish, inspect, and retrieve npm packages and Knowledge Graphs distributed as tarballs or packages from npm registry or GitHub Packages. cli npm-kg-distribution
remote-mount
1 in / 1 out qa-reports
ns-vocabulary · JSON-LD
Namespace vocabulary
Emit the folio namespace as a document that dereferences — one node per class and property, each with an @id, a type, a label and a definition, so a consumer holding only the JSON-LD can resolve any term it meets. cli kg-export 2 in / 1 out qa-reports
open-work-lookup · JSON-LD
Does a path or bean exist on main or any open PR
Answer whether a path, a path prefix or a bean id appears on a repository’s default branch or on the head of any of its open pull requests, and where: the PR number, head branch, first commit author and creation time. A path or prefix matches files at the default branch and files each open PR changes; a bean id matches PR titles, bodies and head branch names. Repositories or PRs that could not be read are listed, so an empty answer is never mistaken for absence. inProcess cli bean-coordination 5 in / 2 out qa-reports
openapi-ingest-rendered · JSON-LD
OpenAPI ingest from a published spec
Ingest the OpenAPI documents a layer’s cat-openapi.config.json lists under rendered by fetching each published spec URL (https), verbatim, with a provenance recording the URL, when it was fetched and the document’s sha256 — a URL has no commit, so the digest pins the bytes. A failed fetch or bytes that are not OpenAPI 3 write nothing and never replace a held copy. cli library-ingestion 2 in / 1 out qa-reports
openapi-ingest-source · JSON-LD
OpenAPI ingest from source
Ingest the OpenAPI documents a layer’s cat-openapi.config.json lists under documents from a local checkout of their repository, read AT a branch, tag or commit (git show), verbatim, with a provenance recording the resolved commit and the document’s sha256. --check verifies the committed copies offline. cli library-ingestion 3 in / 1 out qa-reports
openapi-pages · JSON-LD
OpenAPI viewer
Render every OpenAPI document a layer’s openapi graph holds at the visualiser’s route: a hub, a page per layer listing its documents, and a page per document and per operation, drawn from the document copied beside them. cli graph-rendering 2 in / 1 out qa-reports
pack-tarball · JSON-LD
Package layer into an npm tarball and record binary release
Build an npm tarball (.tgz) of a layer using bun pm pack, compute its size and SHA-256 digest, and record it as a folio-binary-release/v1 state node. cli npm-kg-distribution 6 in / 2 out qa-reports
package-exports · JSON-LD
A package’s exports are its declared, shipped code directories
Derive a package’s package.json exports from the layer declaration beside it: a directory declared with graph typology code (top level or from within) AND named in files is exported as ./<dir>/*.js → ./<dir>/*.ts and ./<dir>/* → ./<dir>/*; a declared code directory that is not shipped (test/) stays private. Keys without * are named entry points, kept as written and checked to point into an exported directory. Without --write it reports each difference; with it, it rewrites exports. Exit 0 in agreement, 1 drift, 2 cannot tell (no declaration, no package.json, or no shipped code directory). cli kg-separation 2 in / 1 out qa-reports
package-release-manual · JSON-LD
Package release by hand
A person follows the package-release skill: computes the bump, writes the CHANGELOG entry, tags <package>-v<version>, and creates the release on the host. The same four steps with nothing to configure. manual npm-kg-distribution 2 in / 1 out qa-reports
package-release-workflow · JSON-LD
Release package via GitHub Actions release workflow
Build, tag, and publish an npm package tarball to GitHub Releases and GitHub Packages registry via workflow_dispatch across any repository. cli npm-kg-distribution 4 in / 2 out qa-reports
pages-index · JSON-LD
Published-paper index page
Write the gh-pages index.html for a built paper: a Paper tab embedding the PDF and, when given, a Visualizer tab, with download links and the build’s branch and commit. cli docs-generation 8 in / 1 out qa-reports
pages-publish · JSON-LD
GitHub Pages publish
Push a built directory to the gh-pages branch, where it is served. How the knowledge graph and its schema reach a URL. cli docs-generation
kg-export
2 in / 1 out qa-reports
pdf-cover · JSON-LD
PDF page raster
Render one page of a PDF to a PNG — the thumbnail a repository listing shows — and print the provenance a catalogue needs to record it as DERIVED: source, digest, page, geometry, renderer. cli asset-extraction 5 in / 1 out qa-reports
pdf-viewer-embed · JSON-LD
Embed a PDF inline in a page
Print the HTML fragment that shows a PDF in the installed viewer: a lazily loaded frame whose address is derived from the page’s own location (so one page works at the site root, under a project base and under a staging preview), plus plain open and download links that work without it. cli pdf-inline-viewer 1 in / 1 out qa-reports
pdf-viewer-install · JSON-LD
Install the inline PDF viewer into a built site
Download the pinned pdf.js release (legacy build), verify its SHA-256, copy the parts a site needs into <site>/assets/vendor/pdfjs/, and add the shim that opens ?src= only for the allowlisted URL prefixes or the site’s own origin. Nothing is committed: the viewer exists only in the built site. cli pdf-inline-viewer 3 in / 1 out qa-reports
pin-actions · JSON-LD
Pin third-party GitHub Actions to full commit SHAs
Rewrite every third-party uses: owner/repo@ref outside the declared staging-only workflows to @<full commit SHA> # <ref>, so Dependabot keeps the pair current. A tag resolves to its PEELED commit (an annotated tag’s object is never pinned), else a branch head, by git ls-remote with argv. A ref that resolves to nothing, or to more than one commit, is refused and reported, never guessed. This repository’s own reusable workflows are first-party and left alone. Idempotent. cli security 1 in / 1 out qa-reports
pin-spdx-license-list · JSON-LD
Snapshot the SPDX License List at its pinned version
Read json/licenses.json and json/exceptions.json from a copy of github.com/spdx/license-list-data at the pinned tag and write cat-harness/external-schemas/spdx-license-list.terminology.json (folio-pinned-terminology/v1): every licence and exception id with its name and deprecated flag — the offline, version-fixed list check:source-licence validates a licence.json id against. cli vocabulary-authority 1 in / 1 out qa-reports
processes-viewer · JSON-LD
Processes viewer
Render the declared BPMN processes as an index page and one page per diagram. cli graph-rendering 3 in / 1 out qa-reports
public-comment-changesets · JSON-LD
Public-comment change-sets
Bring a public review’s change-sets back in line with GitHub: open the primary issue of each change-set somebody has engaged with, link the other issues that discuss it, and re-render each issue’s change-set section from its changesets/CS-*.json record. The nightly half of what the folio’s public-comment workflow does on each issue event (github); check validates the records without writing. cli github action public-comment 2 in / 1 out qa-reports
publish-extracted-statements · JSON-LD
Publish extracted statements with source links
Render each declared page (a folio-statement-page/v1 spec naming folio-extracted-statements/v1 lists) as Markdown: every statement extracted from a library source, verbatim, with its kind and status, linked to its library entry and to the section it came from, labelled in the source’s own unit (“Catch up schedule, slide 7”). A statement whose source does not resolve fails the run and nothing is written. --check fails when a committed page is stale. cli publish-extracted-statements 3 in / 1 out qa-reports
publish-gh-pages · JSON-LD
Publish a built site to gh-pages as a full replace
Lay a built _site onto a gh-pages checkout as a FULL REPLACE and push it, keeping the open pull requests’ staging previews: each attempt re-reads the branch, re-runs restore-staging and keep-shared-payloads against what is on it now, replaces the tree, commits docs(gh-pages): site from <sha> and pushes. A rejected push is rebuilt, never rebased (a rebase would re-apply the replace over a preview a sibling just deployed), up to three attempts. A restore that cannot read the branch stops the publish. A site that matches the branch is a no-op, not a failure. It publishes and does not verify: the workflow’s verify step runs after it. cli render-kg-to-cdn 4 in / 1 out qa-reports
qa-sweep · JSON-LD
QA sweep
Run every registered criterion over the blocks under a path and write a per-block QA sidecar. A sidecar rather than a console report, because a printed verdict cannot distinguish “never checked” from “checked and clean”. cli content-test 4 in / 1 out qa-reports
rail-standalone-pages · JSON-LD
Give every page Jekyll did not lay out the folio-assistant navbar
A LAST pass over the finished site: inject the shared navbar (lib/navbar.ts, the same component the site sidebar draws) into every page that carries none — standalone viewers, generated layer pages, and each IG site built by stage-ig-sites. A page under an igSite layer’s route is railed as THAT layer’s page (its name, mark and graphs); a page may declare its own section (data-fa-visualiser-nav) and its label (fa-visualiser-label). Mount routes are left to the mount pass, and a page that declines (folio-navbar: none) is left bare. cli harness-tiles 4 in / 1 out qa-reports
readme-audit · JSON-LD
Audit README links
Verify every Markdown link in a folio’s README still resolves — relative paths against the tree, repo refs against a real ls-tree, Pages URLs against the publish ref. Writes nothing. inProcess mcp cli docs-generation 3 in / 1 out qa-reports
readme-sync · JSON-LD
Sync generated README sections
Rewrite each generated README region, and only where the README already carries that section’s marker pair. Nothing outside a marked region is touched. The layer’s AGENTS.md is synced too, for the skills:index index only: the skills the layer publishes, or, past 32 KiB, their directories with the full list in SKILLS.md. inProcess mcp cli docs-generation 5 in / 1 out qa-reports
references-extract · JSON-LD
Reference lists to reference nodes
Split each library entry’s reference lists (a References, Bibliography or Further reading section or heading, in the six UN languages and Portuguese) into folio-reference/v1 nodes in the instance’s declared references directory, one per cited work, keyed <doc-id>.<section>.<n>: provenance extracted, derivedFrom the section file, printed the entry verbatim, and only the fields read without guessing (URL, DOI, ISBN, year, access date). The bibliography page lists them as the document’s own bibliography. --check writes nothing and fails when a node is stale. cli l1-document-ingestion 3 in / 1 out qa-reports
references-render · JSON-LD
Reference list from reference nodes
Rewrite each reference-list block of a document (a block whose meta.referenceList is set) from the folio-reference/v1 nodes its cites names, in that order: the node’s printed text verbatim when it has one, otherwise WHO style (or the short Authors. Title. Year. form), as a numbered or bulleted list. The block’s .md is generated whole. --check writes nothing and fails when a list is stale. cli graph-rendering 2 in / 1 out qa-reports
release-please · JSON-LD
release-please (declared, not configured here)
Propose the next version of each package from conventional-commit messages, open a release PR with the CHANGELOG and version bump, and — when that PR is merged — create the tag and GitHub release. Does not publish to a registry. Declared here, not configured: no config file and no tag exist in this repository. cli npm-kg-distribution
package-release
2 in / 1 out qa-reports
remote-mount · JSON-LD
Mount remote harnesses at a pinned commit, only when trusted
Lay down each declared remote mount — a source.remote layer in the root index.config.json, or, in a folio with no index, a remoteMounts entry on its declaration (both at once is refused) — and its dependency closure, from another repository at a full commit SHA. Write the lock, index.lock.json (a legacy <name>.mount-lock.json is read, and renamed onto it when rewritten), and, with an index, regenerate the root .gitignore block that ignores each mount path. It READS the declared mounts and never writes them: a tool that changes a mount writes through writeDeclaredMounts into index.config.json. Before anything is checked out, each mount must pass the trust gate (schemas/kg/mount-trust.ts, rule H8): a person’s consent recorded for THIS pin, or a signature in a declared trust network. No signature verifier exists yet, so a signature alone is could-not-determine and does not mount. Unsigned and unconsented is refused. --staging mounts for a preview and needs neither. A directory id an override or mountDefaults names may be one the layer declares in its <name>.json or one it declares FROM WITHIN a declared directory (skills/skills.json naming voices), resolved at the pin through instanceDirectories; an id in neither is refused at plan time. tools on an override names Tools by id and mounts each one runnable: its definition (<tools dir>/<id>/<id>.ts), its implementation (the inProcess.module, or the script its bun run cat <key> resolves to) and their import closure, walked by bootstrap-tools’ importClosure (the one closure walker), as every declared directory that closure touches — the lock still records directories only. A hydrated mount lays each chosen directory down as its published index.hydrated.jsonld only: read from the pin’s git tree when a commit carries it, otherwise fetched from the pin’s subgraph index package (the subgraph-index-<sha> release asset subgraph-index-package publishes), its tarball checked against its folio-binary-release/v1 record, its manifest against the repository, pin and harness asked for, and each file against its sha256, SubgraphHydratedSchema and the subgraph IRI; the lock records the package’s URL and sha256 so a replay fetches the same bytes. --check compares the disk against the lock and never fetches. When a run CHANGES the lock, it then re-renders the agent surface (skill:commands → .claude/commands/, agent-memory → .claude/agent-memory/ and .agents/) from the checkout root, so a re-pin commit carries the skills and memory the new layers define; a generator that fails fails the run, and --no-agent-surface opts out. cli agent-memory
security
12 in / 1 out qa-reports
render-log · JSON-LD
Record what the publish branch served
Append one entry to the render log on the publish branch: what was published or taken down, when, from which commit, and — for a removal or a retention — WHY. The log is the only place a preview that vanished leaves a trace. cli render-logging 9 in / 1 out qa-reports
render-order · JSON-LD
The render pipeline, in dependency order
Flatten the repository’s renders into the order their needs imply, and optionally run them. Two stages: the current declared state as json/jsonld and the README derived from it are FATAL; the dynamic renderers (viewers, visualisers, doc pages, diagrams) skip and log; the dynamic-state export closes it. A cycle or a missing dependency yields NO order rather than a partial one. inProcess mcp cli docs-generation 1 in / 1 out qa-reports
review-log-watch · JSON-LD
Check a review log and its repository’s issues for changes
The mechanical check a review-log watcher runs on every poll. Given a drafting sheet’s Contributor Feedback values and the folio repository, it reads the repository’s open and closed issues (leaving out pull requests, issues labelled no-sheet and, with --author, issues someone else opened) and compares a digest of them and of the cells people set (Title, Selected Sections, Contributor Feedback, Disposition, Ready for agent, Ready for merge, Dispositions confirmed; on a public-comment log also Committee Feedback, TWG (CS), TWG (human) and Change Set (human)) with the last check’s. Unchanged: it prints {"status":"unchanged"} and exits 0. Changed: it exits 10 and lists what changed (issue opened, updated or closed; a cell’s old and new text), the Contributor Feedback rows to append, one per issue that no row’s Issue cell links yet (Title the issue’s title, the first Contributor Feedback column @login (#N): first paragraph, Issue the issue’s URL; a closed issue’s row marked closed), and closedRows, the rows linking a closed issue, which the watcher hides with their text struck through. A read with no Issue column (a public-comment log’s Review Change Sets) tracks only issues labelled change-set unless --label says otherwise, and appends no rows: each open issue is reported as an issue-unlinked change instead. It writes no sheet and no issue: the watcher appends the rows and does the agent pass. With --accept it stores the new state for the next poll to compare against. cli review-log-watch 8 in / 1 out qa-reports
review-sheet-sync · JSON-LD
Build the review sheet’s Apps Script and a document’s Config tab
Builds Code.gs (--script), the one Google Apps Script every review sheet runs (from review-sheet.gs), and one document’s Config tab rows (--config), filled from the folio’s review/public-comment/sheet-sync.json, its review anchors and the draft’s pdf-line-map.json. The script keeps a public-comment review sheet’s links and joins current: each comment No. links to its comment page; Sections, Page and Line no(s) link to the page of the frozen review draft; Review Change Sets joins its comments, suggested revisions, feedback and dispositions, each No. n linking to its row; and on both tabs Agentic status lists each pull request of the change-set, linked, with its status and what it changed, from the Agent tab. It reads the document’s URLs, tab and column names and data from the sheet’s Config tab (visible, locked to the owner), which the agent writes from these rows. A person pastes the script (one file) into the sheet once (Extensions > Apps Script) and runs installTriggers; from then on the sheet rebuilds itself within about a minute of an edit, with no agent. With "mode": "drafting" in the config it builds the drafting sheet instead: one Contributor Feedback tab where each row is a change set (ids assigned by the script), Sections picked from the document’s table of contents and linked to it, and Issue and Agentic status from the Agent tab. cli review-sheet-script 4 in / 2 out qa-reports
rewrite-moved-paths · JSON-LD
Rewrite paths that still name a moved file’s old layer — never in a record
After a move between two layers lands, find every mention <from>/<prefix>/<rest> in the scanned tree’s tracked files and rewrite it to <to>/<prefix>/<rest> only when the file exists under the destination root and not under the source root. Records are never read: beans, todos, memory, proposals, wireframes, QA results, upstream pins, vocab mappings, provenance figures and terminologies, and test fixtures (DEFAULT_RECORD_GLOBS; the CLI’s --keep <glob> adds more, generated trees among them, which their generators rewrite — a glob is not a shell-safe argument type, so it is not a declared input). Reports by default with file and line; --write applies; --check exits 1 while a rewrite is owed. cli kg-separation 8 in / 1 out qa-reports
schema-docs · JSON-LD
Skill contract reference
Render each skill’s input/output JSON Schema as a browsable Markdown reference page, with an index. The generated pages are committed so they are readable on the forge as well as on the site. cli docs-generation 1 in / 1 out qa-reports
schemas-viewer · JSON-LD
Schemas viewer
Render each declared schema directory as a page per subject layer, over one shared data index. cli graph-rendering 2 in / 1 out qa-reports
search-graph · JSON-LD
Lexical search over the content graph
Find graph nodes, authored or ingested, whose label, title, kind or tags contain a string, with no graph expansion. corpus-search composes this with a neighbour walk. inProcess mcp corpus-grep 4 in / 1 out qa-reports
secrets-check · JSON-LD
Check declared credential needs against workflow use and expiry, by name
Compare each repository’s declared credential needs (credentials/needs.json, profile and registry) with the secrets.NAME expressions its workflows reference (GitHub Actions’ dollar-and-double-brace form, said in words because this text is rendered on a Jekyll page), with the secret names GitHub lists when a token allows it, and with any recorded expiry. NAMES ONLY — no value is ever read. Findings: used-not-declared, declared-not-used, declared-not-present, present-not-declared (reported, never deleted), naming advisories, and expiry (≤30 days minor, ≤7 major, expired critical). Exits 2 when anything could not be determined, which outranks clean; else 1 on a major or critical finding; else 0. cli security 2 in / 1 out qa-reports
security-gate · JSON-LD
Run every existing security check as one named release step
Run the repository’s security checks — workflow injection, secret leaks, lockfile pinning, the toolchain pin, QA reviewer permission, materialised-asset fixity — plus two advisories (dependency advisories, third-party action SHA pinning), each by name as argv. Every check is reported as pass, fail or unknown; a blocking check that fails OR could not be run refuses the release, because could-not-check is never clean. Advisories are reported and never block. Called by prepare-merge before a push that will merge, and by any publish process before it publishes. cli security 1 in / 1 out qa-reports
separation-cutover-check · JSON-LD
Is this repository’s cutover from the monorepo complete? Measured, condition by condition
Run in a separated repository. Six conditions, each decided by an existing gate: mounts (every declared layer on disk at its pin — remote-mount --check), declarations (every path a declaration names resolves — declaration-resolution), monorepo-paths (no wiring still names the monorepo: livesAt on it, a submodule of it, a reusable workflow called from it, a path into its old checkout — the one condition implemented here), import-direction and reference-direction (no layer reaches a layer above it), and site (the site builds from the mounts alone, every link resolving — build-instance-site --no-mount, only with --site). Findings inside a MOUNTED layer are listed and not counted: a defect in a layer pinned from upstream is fixed upstream and re-pinned. Exit 0 complete, 1 a condition failed, 2 a condition could not be determined or was not run — without --site the answer is never 0. cli kg-separation 5 in / 1 out qa-reports
serve-rendering · JSON-LD
Local rendering server
Serve a layer’s renderings over local HTTP with their declared media types. The publication host wherever GitHub Pages is absent, and the only host that can enforce application/ld+json at all. cli serving-renderings 2 in / 1 out qa-reports
serve-rendering-py · JSON-LD
Local rendering server (Python)
Serve a layer’s renderings over local HTTP with their declared media types using Python 3 standard library. The publication host wherever GitHub Pages is absent, and the only host that can enforce application/ld+json at all. cli serving-renderings 2 in / 1 out qa-reports
shared-assets · JSON-LD
Stage a site’s shared asset directories where the host serves them
Write each shared directory of a built site (its assets/… laid out as a built site) to <out>/shared/<dir>/<key>/ with a .shared.json, so sites on one host load one copy instead of one each. A directory already there is left alone. This is the host side of the shared-assets step build-instance-site runs when a checkout declares site.sharedAssets; run on its own, it stages an IG build, or any built site, into a host site root. cli docs-generation 2 in / 1 out qa-reports
sibling-sessions · JSON-LD
Sibling sessions in a window
List the Claude Code sessions that have committed to this repository in a time window, from the Claude-Session: trailer on commits across ALL branches — with each one’s commit count, first and last commit, latest subject, and the branches containing its tip. The session API cannot see a sibling session, so the trailer is the only durable session identity here and a session’s state is INFERRED from its branch; whether a session is still running is not knowable from a checkout and is deliberately not reported. inProcess cli bean-coordination 2 in / 2 out qa-reports
site-build-local · JSON-LD
Build the docs site locally
Build the published site on this machine, so a page can be looked at rather than described: the same Jekyll build pages-publish runs in CI, into a directory of your choosing, and never pushed. The local half of the two site builds: local and GitHub builds are two Tools for one skill. cli docs-generation 1 in / 1 out qa-reports
site-search-index · JSON-LD
Site search index
The just-the-docs search index, assets/js/search-data.json, which the theme writes as part of the Jekyll site build: one entry per page section, searched by every page’s search box. Built implicitly by the build rather than by a command of its own. cli docs-generation 0 in / 1 out qa-reports
site-search-scopes · JSON-LD
Site search scopes
The site search index cut into one index per scope — each declared layer, each target locale, and the platform — plus assets/js/search/manifest.json naming them, so a reader’s search loads its own scope rather than the whole site. Run on the assembled site after the index is written or borrowed. cli docs-generation 1 in / 1 out qa-reports
site-viewer · JSON-LD
Published site viewer
List the published site’s top-level routes and staging previews, with the pages and files under each, as read from the branch that keeps the site — saying so when the branch is empty or could not be read, never drawing zeros in its place. cli graph-rendering 2 in / 1 out qa-reports
skill-docs · JSON-LD
Skill instruction reference
Render the skill instruction bodies — the prose an agent actually loads — as browsable pages with an index, so a reader can see what an agent is told without cloning the repository. cli docs-generation 1 in / 1 out qa-reports
skill-fetch · JSON-LD
Fetch a skill
Load one skill’s instruction body for the agent to follow, from the local packages or an external bundle. inProcess mcp skills-and-tools 2 in / 1 out qa-reports
skill-list · JSON-LD
List skills
Every skill this layer can resolve, with its one-line summary. The entry point AGENTS.md sends an agent to first. inProcess mcp skills-and-tools 0 in / 1 out qa-reports
skill-update · JSON-LD
Edit a skill
Change one existing local skill’s instruction file through a gate: it must resolve as skill-fetch serves it, be tracked in the working repository, not be generated, be granted skill-update to the actor by a policy, be unchanged since it was read, keep its name and parse, and add only links that resolve. A gate that cannot decide refuses. Writes the working tree; the caller commits. inProcess mcp skills-and-tools 9 in / 2 out qa-reports
slice-sqlite · JSON-LD
Per-slice SQLite builder
Build a named slice of a graph (beans, todos, library, or the whole-repo kg) as one SQLite file a browser mounts without parsing it, beside a manifest carrying its sha256, a row-content digest, its row counts and the search block the one search page reads. Heavy text is indexed by a contentless FTS5 and not stored; each row points at a content-addressed payload. --check builds every slice twice, requires one sha256, and requires the row digest read back from the file to equal the one computed from the source. cli kg-export 4 in / 2 out qa-reports
smart-approach-svg · JSON-LD
WHO SMART Guidelines approach chart
Draw the WHO SMART Guidelines approach chart (knowledge layers L1–L5 with their products, and their country adaptations, between the research, localisation and impact bands) as SVG from smart-guidelines-approach-figure/v1 data on a who-iris library entry, such as the product catalogue handbook’s Fig. 5. Every label comes from the data. cli figure-rebuild 2 in / 1 out qa-reports
stage-local · JSON-LD
Publish a folio’s or an IG’s site from the agent’s own checkout — the one agent publish path
Build a repository’s site in the agent’s checkout and push it to the publish branch, as a preview or as the main site, without an Actions runner. Which build is read from the repository: a declared site.build recipe (built by build-instance-site), else the reusable workflow its own workflow files call — folio-staging.yml for a folio, the WHO SMART Base IG’s ghbuild.yml for a FHIR IG, whose build job is planned step by step and run as written. For an IG it then applies the harness chrome over the built output (sync-docs-harness, then rail-standalone-pages --foreign-site --instance <the IG>): every page gains the rail and the icon row #fa-navbar-row, and the IG’s own content is left byte for byte. A root publish is REFUSED when the chrome cannot be applied, and a clean root deploy is refused over any root whose index.html wears the icon row unless the new output does too, so a chromed root is never replaced by bare IG Publisher pages. Every run prints its plan first, and the plan lists the chrome step; --plan stops there, --dry-run composes the commit in a temporary worktree and pushes nothing, --artifact writes a claude.ai Artifact bundle instead. cli github action feature-staging
harness-tiles
10 in / 1 out qa-reports
staging-graph-viewer · JSON-LD
Staging-only graph viewer
Render a kept-but-unpublished workspace graph as one page, for a local build or a staging preview only; a canonical deploy withholds it. cli graph-rendering 2 in / 1 out qa-reports
stakeholder-map · JSON-LD
Stakeholder map
Given the paths a proposed change touches, report which skills change, which roles declare them, and who therefore has a stake in the review. inProcess mcp cli role-model 1 in / 1 out qa-reports
state-viewer · JSON-LD
State graph viewer
Render each declared state graph with a projection as a dashboard page: what the work plan holds, and what state it is in. cli graph-rendering 3 in / 1 out qa-reports
sub-kg-verify-clone · JSON-LD
Verify a separated repository from a fresh clone
Clone a separated sub-KG’s repository (with submodules, and any sibling checkouts it needs) into an empty scratch directory, install, and run its gates: its gates script, else its test script (a person may pass --gate on the command line instead; it is not part of this contract because it is a shell command). Reports green, red or unknown; an empty tree, a failed clone or a repository with no gate is unknown, never green. Catches what a rehearsal inside this checkout cannot, because a rehearsal shares this checkout’s node_modules and environment. Writes only inside the scratch directory. cli sub-kg-lifecycle 5 in / 1 out qa-reports
subgraph-index-package · JSON-LD
Package one pin’s subgraph index files for a hydrated mount
Pack every built index.jsonld and index.hydrated.jsonld under a layer’s <docs>/subgraph/<harness>/ into ONE npm package for one pin — <harness>-subgraph-index@0.0.0-pin.<sha>, with a folio-subgraph-index/v1 manifest giving each file’s sha256 — through pack-tarball (bun pm pack), which also writes the tarball’s folio-binary-release/v1 record. The reusable workflow layer-publish-subgraph-index.yml attaches both to the GitHub release subgraph-index-<sha>, where remote-mount finds them for a hydrated source whose pin’s git tree holds no index file (they are builtAtPublish, so none does). It builds nothing: run subgraph:jsonld first. No subgraph files under the layer is a refusal, never an empty package. cli github action npm-kg-distribution
remote-mount
5 in / 2 out qa-reports
subgraph-readmes · JSON-LD
Directory READMEs from the Knowledge Graph
Write a README for every directory a layer declares, from the declaration and the files themselves: the declared title and description, the Graph Typologies, and one row per file described from the file, with ‘used by’ only where a diagram records it. Renders the Liquid templates in tools/templates/readme/, part of the tools graph, which may include one another with Jekyll-style include tags. Writes only between <!-- kg:subgraph:begin --> and :end; a README without the markers is left alone and reported. Records every missing title, missing or over-long description, absent directory and unmarked README in test/results/subgraph-readmes.qa-results.json. cli github action docs-generation
upload-routes
1 in / 1 out qa-reports
subgraph-resolve · JSON-LD
Resolve a declared subgraph’s content source
Say where a declared subgraph gets its content — the checkout’s own directory, a declared repository branch (with its keying), or a branch family (with its prefix, key and repository) — after the layer config’s subgraphSources override by id, and which layer answered. The same declaredSubgraph resolver the KG export, the publishers and the mount tool use, so a shell sees the answer they act on. cli directory-conventions 3 in / 1 out qa-reports
tabular-csv · JSON-LD
CSV tabular metadata
Read a delimited text file into a folio-tabular-csvw/v1 record: one table, its columns and their CSVW datatypes, each measured or explicitly undetermined. Routing is by content, never by extension: the first rows must agree on one field count of two or more, under exactly one of , ; TAB \|. A CSV’s fac:anchor sheet and cell are a determined null. Governed by the tabular-metadata skill. cli tabular-metadata 3 in / 1 out qa-reports
tabular-xlsx · JSON-LD
Spreadsheet tabular metadata
Read a workbook into a folio-tabular-csvw/v1 record, one CSVW table per rectangular block of cells. Blocks are split on empty rows, then on empty columns, so several tables on one sheet are found, each anchored at its real top-left cell (fac:anchor: sheet, A1 cell, row, column). Dates are read through the number formats in styles.xml, not misread as integer serials. A malformed part is refused rather than half-read. Reads the ZIP with node:zlib and SpreadsheetML with fast-xml-parser. Governed by the tabular-metadata skill. cli tabular-metadata 3 in / 1 out qa-reports
theme-viewer · JSON-LD
Themes viewer
Render every theme in the checkout, the platform’s and each layer’s declared themes, with palette, contrast and art. cli graph-rendering 2 in / 1 out qa-reports
themes-css · JSON-LD
Theme stylesheet
Render the declared theme nodes into the stylesheet the site serves. The nodes are the source: a colour has one home, and light and dark are two valuations of one token set rather than two hand-kept blocks. cli site-presentation-assets 1 in / 1 out qa-reports
tool-coverage · JSON-LD
Which uncovered skills warrant a Tool?
Triage the skills that have no Tool by EVIDENCE rather than by grep: a serviceTask naming it or an I/O contract puts it in tier A, a userTask only in B, a shell block or a declared script in C, and nothing in D. The answer to “which of these still have their mechanism inlined in their prose”. cli code-node-review 0 in / 1 out qa-reports
tool-schema · JSON-LD
Tool node schema
The zod definition of a Tool node — what defineTool accepts — and the published JSON Schema generated from it. cli kg-export 1 in / 1 out qa-reports
tool-types-schema · JSON-LD
Tool I/O type vocabulary
The zod definitions of the shared types a Tool’s inputs and outputs reference by IRI, and the published JSON Schemas those IRIs name: one document per type, tool-types/<Name>.schema.json, plus the combined tool-types.schema.json. cli kg-export 1 in / 1 out qa-reports
tools-viewer · JSON-LD
Tools viewer
Render this layer’s Tool nodes, and the skills each satisfies, as one page. cli graph-rendering 3 in / 1 out qa-reports
transcribe-faster-whisper · JSON-LD
Transcribe audio — faster-whisper (option, not installed)
Transcribe an audio file with faster-whisper (the Whisper models on CTranslate2), from Python. A declared OPTION: not installed in this repository. cli library-ingestion 2 in / 1 out qa-reports
transcribe-vosk · JSON-LD
Transcribe audio — vosk (option, not installed)
Transcribe an audio file with vosk (Kaldi-based), offline, from Python, with small per-language models. A declared OPTION: not installed in this repository. cli library-ingestion 2 in / 1 out qa-reports
transcribe-whisper-cpp · JSON-LD
Transcribe audio — whisper.cpp (option, not installed)
Transcribe an audio file to a timed transcript with whisper.cpp, offline, on CPU. A declared OPTION: not installed in this repository until the first audio upload. cli library-ingestion 2 in / 1 out qa-reports
translation-adjudicate · JSON-LD
Record a translation adjudication
Turn one translation-review issue into the recorded adjudication: read the issues event (never the body from a shell variable), refuse an author who is not an owner, member or collaborator, refuse an item whose catalogue hash changed since the page showed it, then write the translation node’s sign-off, the fuzzy suggestions and an append-only translation-qa attestation, and leave a Markdown summary. Exit 0 recorded, 3 nothing recorded, 2 could not determine. cli translation-adjudication 3 in / 2 out qa-reports
translation-adjudication-workflow · JSON-LD
Translation adjudication workflow (reusable)
The tools layer’s reusable layer-translation-adjudication.yml, called by a layer on issues events: runs once per issue labelled translation-adjudication or titled [translation review] …, mounts the platform’s layers, runs translation:adjudicate on the event, and either commits the records on translation-adjudication/issue-<n> (authored by the issue’s author) and opens a ready PR, or comments why nothing was recorded. It never merges. manual translation-adjudication 3 in / 2 out qa-reports
translation-candidates · JSON-LD
Collect candidate translations from open engines
Ask one or more open, self-hosted machine translators — OPUS-MT (Helsinki-NLP Marian models, CC-BY 4.0, run locally) and/or a self-hosted LibreTranslate (AGPL, Argos models) — for CANDIDATE translations of each empty or fuzzy string, so no freemium API quota is involved. Candidates are appended to <locale>/<catalogue>.candidates.json (translation-candidates/v1) for a reviewer to choose among on the translation review page; a candidate whose placeholders or markup do not round-trip is not recorded. Candidates are never a barrier to having a translation: an EMPTY msgstr gets the first good candidate, marked #, fuzzy; a non-empty msgstr is never overwritten. The translation node records the source (generatedBy: "machine", model, time); official is never set, and a candidate is never a verdict. A catalogue that says DO NOT MACHINE-FILL, or whose node is official, gets candidates but no fill. cli translation-candidate-source
translation-refresh
8 in / 2 out qa-reports
translation-candidates-agent · JSON-LD
Candidate translations — agent model
An agent model proposes a translation for each string of a layer’s catalogues, with the locale term base and the string’s source context in front of it, and appends them to each catalogue’s candidates sidecar with source.kind: agent and the model’s id. Seeds an empty msgstr with the first candidate, fuzzy. manual translation-candidate-source 2 in / 2 out qa-reports
translation-extract · JSON-LD
Extract translatable strings
Segment a folio’s prose into a GNU gettext .pot template, leaving code, math and identifiers untranslated. inProcess mcp translation-manager 3 in / 1 out qa-reports
translation-inject · JSON-LD
Inject translations
Produce a translated copy of a source markdown file from a .po, using the same segmentation the extractor used. inProcess mcp translation-manager 3 in / 1 out qa-reports
translation-platforms · JSON-LD
Exchange catalogues with a translation platform
For Weblate, Crowdin and Transifex: export writes the layout file a platform reads from the layer’s declared translation-sources directory; import turns the .po changes a platform pushed onto the checked-out branch into CANDIDATES, never adjudications. Nothing here talks to a platform or the network. Exit 0 done, 1 --check found drift, 3 import found nothing to rewrite, 2 could not determine. cli translation-candidate-source
translation-manager
7 in / 1 out qa-reports
translation-qa-prune · JSON-LD
Prune round-trip rulings of removed strings
Drop the translation-semantic-roundtrip rulings of strings the catalogues no longer hold, and renumber the rest to catalogue order. Rulings are paired with strings by the hash of their msgid, so removing strings invalidates none and needs no agent re-run; this is the upkeep that follows. It only drops or renumbers: a verdict is never changed or invented, and a ruling for a string the ref’s history never held (one recorded ahead of its PR) is kept. cli translation-manager 2 in / 1 out qa-reports
translation-review-page · JSON-LD
Translation review page
The page a reviewer adjudicates on: review.html beside the translation status page, showing every catalogue × locale with its state (machine, flagged, reviewed, stale, official), each string’s candidates, and the hash of the catalogue’s bytes. Submitting opens GitHub’s new-issue form prefilled with a translation-adjudication/v1 record, or copies the body when it is too long for a URL. manual translation-adjudication 1 in / 1 out qa-reports
translation-roundtrip-record · JSON-LD
Record a round-trip translation verdict
Write a back-translation verdict, produced by a pair of translation agents, into a block’s existing translation-QA sidecar. It records a judgement rather than making one, and refuses where no sidecar exists — a round trip cannot be what decides a block is translated. cli translation-manager 1 in / 1 out qa-reports
translation-signoff · JSON-LD
Sign off a translation
Record a translation as official — who signed off, when, and the hash of the source it was signed against, so a later source edit is detectable. inProcess mcp decision-audit
translation-manager
4 in / 1 out qa-reports
translation-status · JSON-LD
Translation coverage
Per-locale translation coverage, computed from the .po files under the translations directory. inProcess mcp translation-manager 1 in / 1 out qa-reports
translation-status-viewer · JSON-LD
Translation status viewer
Render the translation status of every declared translation-sources directory, per locale, as one page, and beside it the review page (review.html, with assets/translation-status/review.json) a reviewer adjudicates on. cli graph-rendering
translation-adjudication
2 in / 1 out qa-reports
translation-validate · JSON-LD
Validate a translation
Check a .po against its .pot — every msgid present, none obsolete, placeholders preserved. inProcess mcp translation-manager 3 in / 1 out qa-reports
uml-object-model · JSON-LD
Harness object model as PlantUML
Draw the harness object model (Actor, Role, Skill, Process, Task, Todo, Bean, tests, schemas) with every attribute read from the schema behind it. Each relationship names the field that carries it, and the generator refuses to write if that field is gone. cli uml-overview 1 in / 1 out qa-reports
uml-overview · JSON-LD
UML overview per named sub-graph
Draw one UML class diagram per harness and one per named sub-graph it declares, as PlantUML, Mermaid and Graphviz DOT from one model, with every class read from the graph typology’s node schema, and render the PlantUML to the SVG each page shows (needs Java; the check does not). The DOT is the page’s Interactive view, laid out in the reader’s browser by the site’s vendored Graphviz WASM with draggable nodes. A kind with none is drawn as could-not-determine, never as an empty box. cli kg-subgraph-layout
uml-overview
1 in / 1 out qa-reports
upload-url · JSON-LD
Where to drop a file — the queue’s upload URL
Compose the forge URL a person can drop a file at, from the layer’s own declaration: the uploads graph’s directory, resolved against the ROOT its scope names, expressed relative to the repository, and appended to the origin remote as GitHub’s /upload/<branch>/<path>. Every failure is NAMED and no URL is guessed — no declaration, no declared uploads graph, a declared queue absent from disk, no origin remote, and a non-github.com remote each return a reason and a remedy instead. The declared path is relative to the LAYER and a forge URL needs it relative to the REPOSITORY, so pasting the declared path can mint a 404. cli content-acquisition
upload-routes
1 in / 1 out qa-reports
uploads-viewer · JSON-LD
Uploads viewer
Render each declared uploads directory’s intakes as a page per subject layer. cli graph-rendering 2 in / 1 out qa-reports
user-auth · JSON-LD
Who is asking, and what may they do
User authentication and authorization. Asks GitHub for the caller’s login and repository role, maps the role to a gateway actor, and answers from the ODRL policies — for an action, or for a BPMN step. Always states that GitHub’s role covers the whole repository, not a sub-graph, node or query path. inProcess mcp deployment-auth
task-authorization
6 in / 1 out qa-reports
vocab-map · JSON-LD
Apply a vocabulary mapping table
Carry one source record into a target vocabulary by a declared folio-vocab-mapping/v1 table: each source field becomes the predicate its row names, in the table’s order, with a DERIVED target (such as dcterms:title rendering skos:prefLabel) copied from its authoritative one so the two cannot drift. A table is shaped like a FHIR ConceptMap: an existing ConceptMap, R4 or R5, is representable as one without loss (fromConceptMap), and a table can be produced as a ConceptMap (toConceptMap), which returns every loss rather than dropping anything silently. First consumer: glossary-export. inProcess vocabulary-authority 2 in / 1 out qa-reports
voices-viewer · JSON-LD
Voices viewer
Render each declared voices directory as a page per subject layer. cli graph-rendering 2 in / 1 out qa-reports
web-app-build · JSON-LD
Web app build
Build every declared web-app directory in the checkout and publish each at its own route, <locale>/<harness>/apps/<id>/, with no site chrome (every page marked folio-navbar: none), plus an apps/ index in the site’s chrome listing each app with its languages and size. Before building, each locale’s translated UI strings are written from the layer’s translations/<locale>/apps/<id>.po to the app’s i18n.out, so building is the injection. --extract writes the translation templates; --check validates the declarations and translations and builds nothing. cli graph-rendering 3 in / 1 out qa-reports
wireframe-check · JSON-LD
Wireframe check at web and mobile viewports
Render each mid-fidelity wireframe candidate of one topic at a web viewport (1280x800) and a mobile viewport (390x844). For each viewport it records script entries for renders, no-overflow and no-placeholder, each pass or fail with a note. It writes a screenshot per viewport and a report.json carrying the use, topic and surface, and exits 1 on any fail and 2 when the candidate set is not one the use admits. cli wireframe-viewport-check 5 in / 1 out qa-reports
witness-conformance · JSON-LD
Witness conformance report
Check every *.witness.json in the folio’s declared computation-witness directories against the two schemas in schemas/computation-witness.ts: the envelope every witness should meet (a failure is a malformed file) and the producer contract (a failure is a finding against the producer, grouped by the fields at fault). Also lists files that are not strict JSON, which Python’s reader accepts and every other consumer rejects. Report-only: it never edits a witness, which is generator output. cli directory-conventions 3 in / 1 out qa-reports
work-plan-prime · JSON-LD
Prime the work plan
Load the current work plan for this session — the same committed beans store the CLI reads, so a fresh container starts from the plan rather than from nothing. inProcess mcp cli bean-coordination
pending-show
session-intent
todo-manager
0 in / 1 out qa-reports
workflow-complete · JSON-LD
Complete a step
Record an enabled step as done — or supply the facts a decision gateway is computed from — and advance the layer. Refuses a step that is not enabled. inProcess mcp bean-coordination
process-state
8 in / 1 out qa-reports
workflow-gate · JSON-LD
May this step be performed?
Ask before doing work a strict process governs. The content-agnostic processes refuse a step that is not enabled; the per-content-type ones advise. inProcess mcp process-state 4 in / 1 out qa-reports
workflow-list · JSON-LD
List processes
The BPMN processes this layer defines, and the layers currently open against them. inProcess mcp process-state 0 in / 1 out qa-reports
workflow-next · JSON-LD
What is enabled now
The activities a layer may work right now, each with the lane that owns it and the skill that implements it. inProcess mcp process-state 1 in / 1 out qa-reports
workflow-start · JSON-LD
Start a process instance
Open an instance of a process for a subject. Idempotent: an existing layer for the same subject is returned rather than duplicated. inProcess mcp bean-coordination
process-state
4 in / 1 out qa-reports