The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data which could contain malicious instructions designed to influence the agent's code modifications.
Ingestion points: In SKILL.md, the workflow requires the agent to inspect GitHub Actions logs and external check links to identify errors.
Boundary markers: The instructions do not define delimiters or provide specific warnings to the agent to disregard instructions embedded within the logs or linked content.
Capability inventory: The agent has the ability to perform local file edits, push branches to remote repositories, and update Pull Requests (as described in SKILL.md and the 'Codex publication boundary' section).
Sanitization: There is no mechanism described for sanitizing, validating, or filtering the content retrieved from logs or external links before it is used to inform code fixes.
[COMMAND_EXECUTION]: The skill relies on the gh (GitHub CLI) tool for its core workflow.
Evidence: SKILL.md explicitly instructs the agent to execute gh pr checks to retrieve check states and interact with the GitHub platform for pushing branches and managing PRs. This is the intended primary purpose of the skill.

cursor-fix-ci