The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a sequence of shell commands to manage a bug-hunting workspace and generate reports. This includes rm -rf .clawpatch for cleanup, mkdir for directory creation, and the clawpatch tool suite (init, map, review, report). These operations are consistent with the skill's documented purpose.
[PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection by ingesting data from external tool outputs and passing it to subagents for analysis.
Ingestion points: SKILL.md (Step 2) reads results from docs/bugs/${DATE}-bugs.md.
Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when interpolating tool output into subagent prompts.
Capability inventory: The subagents are granted file-read capabilities via the Explore tool to verify bug findings.
Sanitization: No sanitization or validation of the ingested clawpatch output is performed before it is passed to the subagents.

clawpatch-bugs