optimize-agents-md
Pass
Audited by Gen Agent Trust Hub on Feb 28, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses local read-only commands for analysis.
- Evidence: Executes
wc -w AGENTS.mdto count words for token estimation. - Evidence: Executes
grep -l "bun test" AGENTS.md package.jsonto identify redundant information. - Evidence: Suggests using
grep 'interface [A-Z]' src/as a verification pattern for coding style compliance. - [DATA_EXPOSURE]: The skill only accesses standard project documentation and configuration files.
- Evidence: Reads
AGENTS.md,package.json, and source code insrc/. These are non-sensitive project files intended for agent context. - [PROMPT_INJECTION]: No malicious prompt injection or bypass instructions found.
- Evidence: The instructions focus strictly on document compression and formatting without attempting to override system safety or agent constraints.
Audit Metadata