intent-extract
Use during bob's Step 4.5 (after UNIT_TESTED) or evo's INTENT_MAPPED phase (per design §6 state machine) to extract per-component functional intent from a project's source code. Anchor-and-expand on contract-map components plus 1-hop call-graph neighbourhoods; per-component LLM pass at temperature=0; structured YAML output conforming to functional-intent.v1 with closed-set function_class enum, REQUIRED evidence_edges that resolve in static.jsonl, two-arm verification on prose claims; content-hash cached. Requires a bob-issued claim token, heartbeats it every 60s, emits transition requests for INTENT_MAPPED. Skill NEVER writes claim files, NEVER writes the integration ledger, NEVER traverses the call graph beyond 1-hop. CB4-compliant.
소스 정보
- 저장소
- joogy06/agent-foundry
- 최근 소스 활동
- 2026년 7월 31일 11:15
- 감지된 SKILL.md 언어
- 영어
- 스타
- 4
- 포크
- 0
설치 방법
기본적으로 소스를 먼저 확인하는 Prompt가 선택됩니다. 직접 명령으로 전환하거나 로컬 사본을 다운로드할 수도 있습니다.
소스 파일 검토
설치 여부를 결정하기 전에 SKILL.md와 SkillsMP에 표시된 보조 파일을 읽어 보세요.
SKILL.md 표시 중
| name | intent-extract |
|---|---|
| description | Use during bob's Step 4.5 (after UNIT_TESTED) or evo's INTENT_MAPPED phase (per design §6 state machine) to extract per-component functional intent from a project's source code. Anchor-and-expand on contract-map components plus 1-hop call-graph neighbourhoods; per-component LLM pass at temperature=0; structured YAML output conforming to functional-intent.v1 with closed-set function_class enum, REQUIRED evidence_edges that resolve in static.jsonl, two-arm verification on prose claims; content-hash cached. Requires a bob-issued claim token, heartbeats it every 60s, emits transition requests for INTENT_MAPPED. Skill NEVER writes claim files, NEVER writes the integration ledger, NEVER traverses the call graph beyond 1-hop. CB4-compliant. |