work-pr-info
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands, specifically
git logand a custom tooldf-ti, to retrieve information from the repository and issue tracker (SKILL.md). - [INDIRECT_PROMPT_INJECTION]: The skill processes data from untrusted external sources like commit messages and issue descriptions, creating a surface for indirect prompt injection.
- Ingestion points: Data is ingested from
git logoutput, branch names, and the output of thedf-ticommand which fetches issue descriptions (SKILL.md). - Boundary markers: Absent. No explicit delimiters or instructions are provided to the agent to separate external content from its own instructions.
- Capability inventory: The agent has the capability to execute shell commands (
git,df-ti) and write files to the local file system (<root>/tmp/) (SKILL.md). - Sanitization: Absent. There is no mention of sanitizing or validating the content retrieved from issues or commit logs before processing.
Audit Metadata