Roles scenarios

Derived: every role a declared `scenarios` graph defines, with the description it declares. Sub-graph cat-harness/scenarios.

This is an index, not the documentation. It says what exists and what each artefact declares about itself. What a process is for, when you would be in it, and what it is not, is authored beside it and reuses these entries rather than restating them — see the auto-docs skill.

folio-assistant-core-scenarios folio-assistant-core/scenarios 3
folio-assistant-sci-scenarios folio-assistant-sci/scenarios 3
scenarios cat-harness/scenarios 47
smart-base-scenarios smart-base/scenarios 1

Filter this table

Artefact	What it declares about itself
`Adjudicator` cat-harness/scenarios/roles.json#adjudicator	Settles a criterion whose reviewer entries disagree, and records WHY.
`Administrator` cat-harness/scenarios/roles.json#administrator	Administers the role graph itself: who exists, what they may act as, and what they may do.
`Attestation service` cat-harness/scenarios/roles.json#attestation-service	Signs a QA report's run hashes over an API.
`Author` cat-harness/scenarios/roles.json#author	Writes and revises folio content.
`Authoring agent` cat-harness/scenarios/roles.json#authoring-agent	The session agent doing the work.
`Board renderer` cat-harness/scenarios/roles.json#board-renderer	Draws a board and performs what the reader has already decided: reads each kind's declared zoom threshold and the layout layer, swaps a card for its avatar, projects an open window onto the canvas, and carries out a con…
`Build pipeline` cat-harness/scenarios/roles.json#build-pipeline	Validates, renders and publishes, and watches whether that actually worked.
`Business analyst` cat-harness/scenarios/roles.json#business-analyst	Turns a stated need into a modelled requirement.
`Certifier` cat-harness/scenarios/roles.json#certifier	Takes the certification decision on ONE test report: applies the DMN the plan's `exitCriteria` names (computed from facts, never asserted) and records the decision as a judgement under the declared `attestations` graph,…
`Clinical / scientific SME` cat-harness/scenarios/roles.json#clinical-sme	Reviewer whose judgement is about the subject matter rather than the artefact.
`Code reviewer` cat-harness/scenarios/roles.json#code-reviewer	Judges the knowledge graph's CODE nodes — Tool definitions in the `tools` graph and schema definition nodes under `schemas/`.
`Corpus` cat-harness/scenarios/roles.json#corpus	The versioned content store and the L1 source knowledge graph.
`Docs authoring agent` cat-harness/scenarios/roles.json#docs-authoring-agent	The authoring agent producing what a person reads ABOUT the instance — generated references, the glossary, sticky notes, theme art and deployment documentation.
`Editor` cat-harness/scenarios/roles.json#editor	Author plus the authority to accept a change into the corpus.
`Editorial authoring agent` cat-harness/scenarios/roles.json#editorial-authoring-agent	The authoring agent in the editorial lane: auditing a block's prose, structure and sourcing rather than writing new subject matter.
`External registry` cat-harness/scenarios/roles.json#external-registry	Trusted registries and observatories outside this instance's control.
`Feedback provider` cat-harness/scenarios/roles.json#feedback-provider	Offers a finding on content or code — a script checker, a review agent, a reader, an SME.
`FHIR modeller` cat-harness/scenarios/roles.json#fhir-modeller	Authors the L3 profiles, extensions and examples an IG is built from.
`IG publisher + validator` cat-harness/scenarios/roles.json#ig-publisher-service	The JVM services an IG build shells out to.
`Ingestion engine` cat-harness/scenarios/roles.json#ingestion-agent	Runs unattended over a source corpus.
`Integration watcher` cat-harness/scenarios/roles.json#integration-watcher	Watches the default branch and open branches for newly-landed work that an existing audit axis should be re-run over.
`Lean toolchain` cat-harness/scenarios/roles.json#lean-toolchain	Elaborates and checks the formal half of a paper folio.
`Librarian` cat-harness/scenarios/roles.json#librarian	Acquires a source and files it.
`Activity log` cat-harness/scenarios/roles.json#log	The agent activity log.
`Merge steward` cat-harness/scenarios/roles.json#merge-steward	Stewards writes to ONE watched ref -- a long-lived branch several processes write and something downstream consumes serially.
`Narrative reviewer` cat-harness/scenarios/roles.json#narrative-reviewer	Judges PROSE — register and voice, the editorial dependencies a reader needs, and translation.
`Platform authoring agent` cat-harness/scenarios/roles.json#platform-authoring-agent	The authoring agent working on how this instance PUBLISHES itself — the MCP projection, its contract, and the rendering endpoints.
`Programme manager` cat-harness/scenarios/roles.json#programme-manager	Owns what gets worked on and holds release authority.
`Publication manager` cat-harness/scenarios/roles.json#publication-manager	Takes an authorised release to its published form and is accountable for what is live.
`Publish target` cat-harness/scenarios/roles.json#publish-target	GitHub Pages, or wherever the folio is served from.
`QC reviewer` cat-harness/scenarios/roles.json#qc-reviewer	Reviewer with the publication quality gate.
`Review coordinator` cat-harness/scenarios/roles.json#review-coordinator	Organises a review without deciding it.
`Reviewer / SME` cat-harness/scenarios/roles.json#reviewer	Reads a change and judges it.
`Session coordinator` cat-harness/scenarios/roles.json#session-coordinator	The authoring agent managing the SESSION rather than the content: what this session intends, what is already in flight elsewhere, and what to pick up next.
`Session record` cat-harness/scenarios/roles.json#session-record	A session's context — who is acting, which instances are open, what it waits on.
`Sibling session` cat-harness/scenarios/roles.json#sibling-session	Another agent or human working the same store.
`Stakeholder` cat-harness/scenarios/roles.json#stakeholder	Signs off that a requirement is met.
`Terminologist` cat-harness/scenarios/roles.json#terminologist	Owns coded concepts and value sets — the vocabulary a DAK is expressed in.
`Test requester` cat-harness/scenarios/roles.json#test-requester	Asks for a system under test to be certified against a named `test-plan/v1` (plan id and version).
`Tester` cat-harness/scenarios/roles.json#tester	Runs a `test-plan/v1` against a system under test: resolves the plan, binds its test data (fixed and reviewed, or generated from template + params + seed, per `vm6m`), executes every case or skips it with a reason, and…
`Translation adjudicator` cat-harness/scenarios/roles.json#translation-adjudicator	Adjudicator who settles a disagreement between machine and human translation.
`Translation coordinator` cat-harness/scenarios/roles.json#translation-coordinator	Decides what is translated, into which languages, and when a translation is accepted.
`Human translator` cat-harness/scenarios/roles.json#translator	Produces the target-language text.
`Untainted checker` cat-harness/scenarios/roles.json#untainted-checker	Re-executes a SAMPLE of a plan's cases from the plan and the data alone — never the tester's verdicts — and reports whether it agrees.
`User` cat-harness/scenarios/roles.json#user	The person the assistant is working for.
`Validation pipeline` cat-harness/scenarios/roles.json#validation-pipeline	The mechanical half of the HCI validation gate.
`Work plan (beans)` cat-harness/scenarios/roles.json#work-plan	The shared store.