triage
Warn
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to reproduce reported bugs from reporter steps and run "relevant tests or commands" from PR diffs. This involves executing arbitrary code provided by untrusted external contributors without specific instructions for sandboxing or safety validation.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted content from issue/PR bodies and comments to influence triage logic and brief generation.\n
- Ingestion points: GitHub issue and PR bodies, comments, and code diff content as referenced in SKILL.md.\n
- Boundary markers: Absent; there are no instructions to use delimiters or to ignore instructions embedded within the processed external data.\n
- Capability inventory: The agent can execute shell commands, perform file system writes (e.g., to the
.out-of-scope/directory), and perform network operations via issue tracker commenting.\n - Sanitization: Absent; no validation, filtering, or escaping of external content is specified before it is processed by the agent.
Audit Metadata