provenance
Pass
Audited by Gen Agent Trust Hub on Apr 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes shell commands including
grep,find, andbasename, along with specialized local utilitiesaoandcassto identify metadata and search session history. - [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it reads and processes the content of local artifact files which could contain instructions.
- Ingestion points: Knowledge artifacts are read from
.agents/learnings/and.agents/patterns/as seen inSKILL.mdSteps 1 and 2. - Boundary markers: Absent; there are no clear delimiters used to distinguish data from instructions within the processed files.
- Capability inventory: The skill has access to
Bash,Read,Grep, andGlobtools, which are used to search and report on artifact history. - Sanitization: No sanitization is performed on the file content before the agent parses it for provenance metadata.
Audit Metadata