The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted content from GitHub issues and repository documentation to drive code changes and command execution.
Ingestion points: The skill reads repository contents using gh api and processes issue descriptions to determine implementation strategies in SKILL.md.
Boundary markers: Absent. The instructions do not specify the use of delimiters or 'ignore embedded instructions' warnings when handling content from external GitHub repositories.
Capability inventory: The skill utilizes shell access for git operations, gh CLI commands, and running arbitrary test suites such as pytest, npm test, and cargo test.
Sanitization: Absent. There is no explicit sanitization or validation of the content retrieved from external sources before it influences the agent's logic.
[COMMAND_EXECUTION]: The skill dynamically identifies and executes development tools and test suites based on configuration instructions found in external repository metadata.
Evidence: Instructions in SKILL.md include running python -m pytest tests/, npm test, or cargo test as found in the target repository's README or CONTRIBUTING.md documentation.

oss-contribute