The Agent Skills Directory

[PROMPT_INJECTION]: Indirect prompt injection surface identified. The skill ingests untrusted data from the repository being audited (including source code and documentation) via search and read tools.
Ingestion points: Entire repository file structure accessed through glob, read, grep, and ast_grep_search (SKILL.md).
Boundary markers: Absent. The instructions do not define clear delimiters or provide warnings for the agent to ignore instructions embedded within the codebase being audited.
Capability inventory: High. The skill utilizes bash for system commands and task() for spawning sub-agents.
Sanitization: Absent. File contents are processed directly to identify patterns without prior filtering or validation.
[COMMAND_EXECUTION]: The skill uses the bash tool to perform repository status checks and health assessments.
Evidence: Runs git log to identify file churn, npm audit to check for dependency vulnerabilities, and bun test to evaluate test suite health.
[EXTERNAL_DOWNLOADS]: Mentions an optional external integration for enhanced analysis capabilities.
Evidence: References the CodeGraph repository at https://github.com/colbymchenry/codegraph as an optional MCP tool for symbol and call graph analysis.

tech-debt-audit