The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute git remote get-url origin for repository detection and runs a local script scripts/vapa-review.sh to fetch issue data. The script is well-structured and uses safe practices like variable quoting.
[EXTERNAL_DOWNLOADS]: The skill fetches issue content (body and comments) from GitHub using the official gh CLI. While this involves network access to a well-known service, the data retrieved is considered untrusted as it can be modified by external users.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection because it ingests and processes untrusted data from GitHub issues to perform its analysis.
Ingestion points: The skill reads the full body and all comments of a GitHub issue via the fetch_context function in scripts/vapa-review.sh.
Boundary markers: The instructions in SKILL.md do not include explicit delimiters or 'ignore embedded instructions' warnings for the fetched content.
Capability inventory: The skill has access to Bash (restricted to specific commands), Read, and Write tools.
Sanitization: No sanitization or validation of the fetched issue text is performed before the agent evaluates it against the review criteria.

vapa-review