The Agent Skills Directory

[SAFE]: The skill operates locally by reading candidate documents and writing a merged result to the docs/threads/ directory. No network exfiltration or unauthorized file access was detected.\n- [NO_CODE]: The skill is composed entirely of natural language instructions; no scripts, binaries, or external dependencies are included.\n- [SAFE]: Evaluation of potential indirect prompt injection: 1. Ingestion points: Candidate Markdown drafts (SKILL.md). 2. Boundary markers: Absent; drafts are read end-to-end. 3. Capability inventory: Filesystem writes to docs/threads/. 4. Sanitization: The instructions require the agent to restate and fold content, which limits the risk of raw instruction execution from draft content.

merge-artifacts