The Agent Skills Directory

[SAFE]: The skill instructions do not contain any malicious patterns, obfuscation, or unauthorized access attempts.
[PROMPT_INJECTION]: The skill defines a workflow that processes untrusted external data, which is a surface for indirect prompt injection.
Ingestion points: SKILL.md Step 1 (Gather context) fetches content and comments from an external issue tracker.
Boundary markers: Absent; the skill does not instruct the agent to use delimiters or ignore potential instructions within the fetched data.
Capability inventory: SKILL.md (Steps 2 and 5) allows the agent to explore the codebase and create new issues.
Sanitization: No verification or sanitization of the input data from the tracker is implemented.

to-issues