Create an operating checklist
Turn a recurring process into a clear checklist with ownership and exception handling.
5,499,623 AI resources indexedWorking toward the world’s largest AI resource index.
Build or rebuild the SDLC artifact index for agent-navigable discovery
--- namespace: aiwg name: build-artifact-index-4927c4faaf6e platforms: [all] description: Build or rebuild the SDLC artifact index for agent-navigable discovery commandHint: argumentHint: '[--force] [--verbose] [--interactive] [--guidance "text"]' allowedTools: 'Read, Write, Glob, Grep, Bash' model: haiku category: documentation-tracking modelRole: efficiency modelTier: economy --- # Build Artifact Index ## Task Build or refresh the artifact index that powers `aiwg index query`, `aiwg index deps`, and `aiwg index stats`. This is the entry point for ensuring agents can discover and navigate project artifacts efficiently. ## Behavior ### Primary: Use `aiwg index build` ```bash # Standard rebuild (incremental) aiwg index build # Full rebuild (ignore cache) aiwg index build --force # With progress output aiwg index build --verbose # Build every configured graph aiwg index build --all ``` Report what changed: new artifacts indexed, updated, unchanged, and removed counts. If the operator explicitly asks for Fortemi Core cache refresh after a successful build, run: ```bash aiwg index sync aiwg index status --json ``` Fortemi Core is the default artifact search backend after sync. Keep the legacy local backend available behind `--backend local` during the phase-out window. ### Archived Query Answers The index includes files in `.aiwg/working/answers/` automatically. These are answers persisted by the `archive-answer` skill and appear in query results and dependency graphs alongside permanent artifacts. Their `status: working` frontmatter field signals they are candidates for promotion to a permanent destination (e.g., `.aiwg/architecture/`, `.aiwg/reports/`). Use the index to find stale working answers that should be promoted or discarded. ### Fallback: Manual Digest Generation If `aiwg index build` is not available (CLI not installed), fall back to manual scanning: 1. Walk `.aiwg/` for all artifact files 2. Extract titles and short summaries from frontmatter/headings 3. Write digest files to `.aiwg/working/digests/` 4. Generate or update `_index.yaml` with paths, summaries, and timestamps ### Post-Build Verification After building, verify the index is healthy: ```bash aiwg index stats --json aiwg index status --json ``` Report total artifacts indexed, coverage by phase, and any orphaned artifacts. When a Fortemi Core cache exists, also report whether it is built, stale, and how many items it contains. ## When to Run - After creating new SDLC artifacts - After modifying existing artifact references - At the start of a new SDLC phase - When `aiwg index query` returns stale results ## References - @$AIWG_ROOT/agentic/code/frameworks/sdlc-complete/rules/artifact-discovery.md — Agent protocol rules - @$AIWG_ROOT/agentic/code/frameworks/sdlc-complete/skills/artifact-lookup/SKILL.md — Query skill - @$AIWG_ROOT/src/artifacts/cli.ts — CLI implementation--- namespace: aiwg name: build-artifact-index-4927c4faaf6e platforms: [all] description: Build or rebuild the SDLC artifact index for agent-navigable discovery commandHint: argumentHint: '[--force] [--verbose] [--interactive] [--guidance "text"]' allowedTools: 'Read, Write, Glob, Grep, Bash' model: haiku category: documentation-tracking modelRole: efficiency modelTier: economy --- # Build Artifact Index ## Task Build or refresh the artifact index that powers `aiwg index query`, `aiwg index deps`, and `aiwg index stats`. This is the entry point for ensuring agents can discover and navigate project artifacts efficiently. ## Behavior ### Primary: Use `aiwg index build` ```bash # Standard rebuild (incremental) aiwg index build # Full rebuild (ignore cache) aiwg index build --force # With progress output aiwg index build --verbose # Build every configured graph aiwg index build --all ``` Report what changed: new artifacts indexed, updated, unchanged, and removed counts. If the operator explicitly asks for Fortemi Core cache refresh after a successful build, run: ```bash aiwg index sync aiwg index status --json ``` Fortemi Core is the default artifact search backend after sync. Keep the legacy local backend available behind `--backend local` during the phase-out window. ### Archived Query Answers The index includes files in `.aiwg/working/answers/` automatically. These are answers persisted by the `archive-answer` skill and appear in query results and dependency graphs alongside permanent artifacts. Their `status: working` frontmatter field signals they are candidates for promotion to a permanent destination (e.g., `.aiwg/architecture/`, `.aiwg/reports/`). Use the index to find stale working answers that should be promoted or discarded. ### Fallback: Manual Digest Generation If `aiwg index build` is not available (CLI not installed), fall back to manual scanning: 1. Walk `.aiwg/` for all artifact files 2. Extract titles and short summaries from frontmatter/headings 3. Write digest files to `.aiwg/working/digests/` 4. Generate or update `_index.yaml` with paths, summaries, and timestamps ### Post-Build Verification After building, verify the index is healthy: ```bash aiwg index stats --json aiwg index status --json ``` Report total artifacts indexed, coverage by phase, and any orphaned artifacts. When a Fortemi Core cache exists, also report whether it is built, stale, and how many items it contains. ## When to Run - After creating new SDLC artifacts - After modifying existing artifact references - At the start of a new SDLC phase - When `aiwg index query` returns stale results ## References - @$AIWG_ROOT/agentic/code/frameworks/sdlc-complete/rules/artifact-discovery.md — Agent protocol rules - @$AIWG_ROOT/agentic/code/frameworks/sdlc-complete/skills/artifact-lookup/SKILL.md — Query skill - @$AIWG_ROOT/src/artifacts/cli.ts — CLI implementation
Imported from the listed AgentSkill.sh category. Complete supplied package files and notices are retained. Required external tools and credentials are not installed. No bundled code was executed during import. Individual agent task behavior has not been tested.