bug-autopsy
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill workflow requires the agent to execute a local Python-based linter script (
assets/postmortem_lint.py) to validate the structure and content of the generated post-mortem. This script uses only standard libraries and is designed for offline, deterministic validation. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and process data from external sources such as git history, CI logs, and issue tracker threads to reconstruct incident timelines, creating a surface for potential injection via malicious log entries or issue comments.
- Ingestion points: Step 2 of the workflow in
SKILL.mdspecifies readinggit log,git blame, CI runs, logs, and issue/PR threads. - Boundary markers: The skill does not define formal prompt delimiters for external content, but it requires that all entries include evidence citations and that inferences are explicitly labeled as such.
- Capability inventory: The skill can write files to the local filesystem and execute a local linter script via the shell.
- Sanitization: The skill implements a deterministic structural linter (
assets/postmortem_lint.py) which acts as a validation layer, ensuring required sections and checkbox formats are present before completion.
Audit Metadata