swe-post-mortem
Pass
Audited by Gen Agent Trust Hub on Jul 6, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill has an inherent surface for indirect prompt injection because it accepts and processes untrusted user-provided data, such as bug symptoms and root cause descriptions. This data is interpolated into a formal report that could potentially be executed as instructions by other automated systems or influence the agent's behavior during report generation.
- Ingestion points: User inputs regarding the bug fix details (repro steps, root cause, validation data) (SKILL.md).
- Boundary markers: Not present; the skill does not explicitly use delimiters to isolate user-provided data from its instructions.
- Capability inventory: Can perform API requests to the JIRA platform (
POST /rest/api/3/issue/<KEY>/comment). - Sanitization: No explicit sanitization or input validation logic is defined for the external content.
- [COMMAND_EXECUTION]: The skill includes instructions to interact with the JIRA REST API to post finalized records. This network interaction is a core functional component of the skill's purpose. The risk is significantly mitigated by a mandatory sign-off requirement, ensuring that the human user reviews and approves the content and destination before any network call is initiated.
Audit Metadata