The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on executing system commands through the GitHub (gh) and GitLab (glab) CLI tools, along with jq for data processing. These are standard operations for development-oriented tasks.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection as it ingests untrusted data from external sources, specifically CI job logs and PR check descriptions. Ingestion points: Command outputs from gh pr checks (SKILL.md, references/ci-queries.md), gh run view (SKILL.md, references/ci-queries.md), and glab ci trace (SKILL.md, references/ci-queries.md). Boundary markers: The instructions do not define specific delimiters to separate untrusted data from the agent's instructions. Capability inventory: The agent has capabilities to execute shell commands including gh, glab, and jq (SKILL.md). Sanitization: There is no evidence of sanitization or filtering of the content retrieved from logs or status descriptions.

git-ci