The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including bug reports, repository instructions, branch diffs, and execution logs. Malicious instructions could be embedded in these sources to attempt to influence the agent's behavior during the triage process.
Ingestion points: Repository instructions, diffs, error logs, test outputs, and user-provided bug reports (processed in SKILL.md and references/bug-report-template.md).
Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore potential instructions embedded within the logs or repository data.
Capability inventory: The skill utilizes file read/write operations for diagnosis and fixing, along with command execution for reproduction and verification (as seen in the Reproduce and isolate and Fix when authorized sections of SKILL.md).
Sanitization: The instructions do not define specific sanitization or validation steps for the content of logs or external data before processing.

bug-triage