The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes content from external sources that can be controlled by third parties.
Ingestion points: The skill fetches PR conversation comments and review threads using scripts/fetch_comments.py, and retrieves GitHub Actions logs via scripts/inspect_pr_checks.py.
Boundary markers: There are no explicit delimiters or boundary markers used in the fetching scripts to isolate untrusted data from the agent's internal instructions.
Capability inventory: The skill has the capability to modify the codebase, perform git commit, git push to remote repositories, and create or update PRs using gh pr create.
Sanitization: The skill does not implement sanitization or filtering of the text content retrieved from GitHub before it is presented to the agent for processing.
[COMMAND_EXECUTION]: The skill relies on bundled Python scripts that execute git and gh (GitHub CLI) commands through the subprocess module. These commands are necessary for the skill's intended functionality of PR management and CI log analysis. The scripts use list-based arguments for command execution, which is a secure practice that prevents shell injection vulnerabilities.

github-ci-fix