메뉴
Format and upload session wrap-up to NotebookLM for persistent semantic memory
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Execute 9-phase spec-first development workflow (Issue → Spec → TDD → Code → Gen → Seal → Verify → Land → Learn)
Execute tri commands (gen, test, verify, seal, verdict) for spec-first development
This skill should be used when user asks to "tri skill begin", "PHI LOOP", "edit .t27 spec", "seal hash", "tri gen", "tri test", "tri verdict", "tri experience save", "tri notebook query", "tri notebook wrapup", or any task requiring spec-first development in the t27 Trinity S³AI project. Implements the canonical PHI LOOP workflow following constitutional laws with NotebookLM-backed semantic memory.
This skill should be used when user asks to "tri skill begin", "PHI LOOP", "edit .t27 spec", "seal hash", "tri gen", "tri test", "tri verdict", "tri experience save", "tri notebook query", "tri notebook wrapup", or any task requiring spec-first development in the t27 Trinity S³AI project. Implements the canonical PHI LOOP workflow following constitutional laws with NotebookLM-backed semantic memory.
| name | wrap-up |
| description | Format and upload session wrap-up to NotebookLM for persistent semantic memory |
| version | 1.0.0 |
| author | Trinity S3AI Framework |
Upload session summaries to NotebookLM for cross-session memory persistence.
.trinity/ state files/wrap-up "Session completed Phi Loop iterations for Ring-071"
Or with full details:
/wrap-up --summary "Implemented NotebookLM backend" \
--decisions "Used notebooklm-py SDK with cookie auth" \
--files "contrib/backend/notebooklm/*.py" \
--steps "Run integration tests"
This skill uses the t27 spec-first approach:
specs/automation/wrapup-auto.t27contrib/backend/notebooklm/wrapup_auto.pyDirect invocation (for debugging):
.trinity/notebooklm-venv/bin/python3 \
contrib/backend/notebooklm/wrapup_auto.py \
--summary "Session summary" \
--session-id "$(git rev-parse --short HEAD)" \
--decisions "Key decisions" \
--files "file1.py,file2.py" \
--steps "Next steps"
notebooklm login (stores in ~/.notebooklm/storage_state.json)--notebook flag (default: "t27-QUEEN-BRAIN")~/.notebooklm/ — browser profile, storage stateSetup Commands:
python -m venv .trinity/notebooklm-venv
.trinity/notebooklm-venv/bin/pip install notebooklm-py
notebooklm login # Authenticate via browser (one-time)
Returns source ID and notebook ID for verification:
{
"notebook_id": "...",
"source_id": "...",
"uploaded_at": "2026-04-08T..."
}