The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local shell script bash scripts/pr-lint.sh to validate the formatting and content of commit messages against the provided rules.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input in the form of git diff output. * Ingestion points: git diff main...HEAD --stat and code hunks (SKILL.md). * Boundary markers: Absent. * Capability inventory: Execution of bash scripts/pr-lint.sh (SKILL.md). * Sanitization: Absent. The vulnerability surface is inherent to the skill's primary function of summarizing code changes.

pr-writing