grill-with-docs
Pass
Audited by Gen Agent Trust Hub on Jul 19, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, suspicious command execution, or data exfiltration attempts were detected. The skill's behavior is consistent with its stated purpose of documentation management.
- [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection because it is instructed to "explore the codebase" to resolve questions. This process involves the agent ingesting untrusted data from project files which could theoretically contain instructions intended to subvert the agent's logic.
- Ingestion points: Local codebase files accessed during automated exploration (SKILL.md).
- Boundary markers: Absent; the instructions do not specify using delimiters or "ignore instructions" wrappers for the explored code.
- Capability inventory: The agent is authorized to write to the local filesystem to create or update documentation files (CONTEXT.md, docs/adr/).
- Sanitization: Absent; there is no mention of filtering or validating the content read from the codebase before it is processed or used to update documentation.
Audit Metadata