plan-status
Pass
Audited by Gen Agent Trust Hub on Mar 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns were detected. The skill's operation is consistent with its stated purpose of assessing plan progress.
- [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface because it processes untrusted data from project documentation files to perform its assessment. However, this is associated with the primary purpose of the skill and is assessed as safe in this context.
- Ingestion points: Local documentation files at 'docs/plans/*.md' and output from 'mcp__goldfish__recall' and 'mcp__goldfish__plan'.
- Boundary markers: Absent; the agent is instructed to read and interpret file content directly without specific isolation delimiters.
- Capability inventory: The skill has read access to the local filesystem and the ability to update plans via the 'mcp__goldfish__plan' tool.
- Sanitization: No explicit sanitization of the input content is performed.
Audit Metadata