Subgraph
https://litlfred.github.io/folio-assistant/cat-harness/directory/sdlc
JSON-LD · in graph cat-harness.jsonld ·
| name | sdlc |
|---|---|
| path | sdlc/ |
| contentSource | {"kind":"directory","declaredIn":"default"} |
| holdsGraph | cat-harness https://litlfred.github.io/cat-harness/v0/ns/graphTypology/cat-harness |
| description | The software-delivery sub-graph — a NAMED SUB-GRAPH of the harness on the pattern of openapi/, and OPT-IN: off in every layer, this one included, unless that layer lists cat-harness.sdlc in its own enableSubgraphs (owner, 2026-11-03, rulings Q12 and Q18: named folders, a clean break, no redirects; cat-harness#991 step 23.2). It holds the development processes (branch, review, merge, gates, QA, publication, release) and the skills that teach them (sdlc-core, spec-kit; crdm, the requirements methodology, moved to the always-on skills/process/crdm/ on owner card Q-CORE-SKILLS A, 2026-11-12); the graphs inside it — sdlc-skills, sdlc-processes and sdlc-tools — are declared FROM WITHIN, in sdlc/graph.json (#980). Its processes and skills use the work plan, so it requires cat-harness.work-plan: a layer enabling it gets the work plan too (owner, 2026-11-13: a requirement is implied). Its cache.bpmn runs incremental-render, which platform-ui holds, so it requires cat-harness.platform-ui as well (owner card, 2026-11-13: move diagrams into groups that require what they name). Off, the material is left out of what an agent is given (skills, Tools, a mount's default list), so a content layer's knowledge graph does not teach the agent the harness's own delivery process (proposal §4.13). |
| readmePath | sdlc/README.md |
| fileCount | 0 |
| inSubgraph | sdlc https://litlfred.github.io/folio-assistant/cat-harness/directory/sdlc |