The Agent Skills Directory

[SAFE]: The skill's primary function is to summarize existing project data into natural language updates. No malicious patterns, obfuscation, or unauthorized data transmission were found.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes untrusted workspace files and raw logs without explicit boundary markers or sanitization. However, the risk is evaluated as safe because the skill lacks critical capabilities such as network exfiltration or system command execution. Ingestion points: Reads from project metadata, rules, README.md, raw daily logs, and infrastructure records via the agent_mcp filesystem (SKILL.md). Boundary markers: None identified in the prompt instructions to isolate data from instructions. Capability inventory: Restricted to reading files and generating text; no network tools or shell execution patterns detected. Sanitization: No explicit validation or filtering of ingested file content.

weekly-meeting