setup-agent-md
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes the contents of existing
AGENTS.mdandCLAUDE.mdfiles. These files are external to the skill and could theoretically contain instructions designed to influence the agent's behavior. The skill mitigates this by requiring explicit user confirmation before any modifications and by treating existing content as intentional user-authored instructions.\n - Ingestion points: Root-level
AGENTS.mdandCLAUDE.mdfiles.\n - Boundary markers: The skill does not use specific markers to delimit the ingested text from its own instructions.\n
- Capability inventory: File reading and writing (restricted to instruction files) and execution of the
git statuscommand.\n - Sanitization: No sanitization is performed on the content of the instruction files.\n- [COMMAND_EXECUTION]: The skill instructs the agent to run
git statusto verify the state of the repository before suggesting changes. This is a standard, non-malicious command for environment inspection.
Audit Metadata