The Agent Skills Directory

[COMMAND_EXECUTION]: Executes shell commands including git diff and search utilities like grep or ripgrep to perform repository analysis and identify patterns in the code base.- [DATA_EXFILTRATION]: Accesses files and code blocks that may contain sensitive credentials, tokens, and private keys. The skill incorporates a safety constraint requiring all matches to be masked (first 4 characters only), which mitigates the risk of full credential exposure in chat history or logs.- [PROMPT_INJECTION]: Ingests and processes untrusted data from the repository's file system and git history, creating a surface for indirect prompt injection.
Ingestion points: File content retrieved via git diff and repository file scans (SKILL.md).
Boundary markers: Absent; the skill does not define clear delimiters to separate untrusted file content from the agent's operational instructions.
Capability inventory: Shell command execution (git, grep, ripgrep) and file system access.
Sanitization: Limited to masking the specific matched secret strings; no general sanitization of the surrounding file context is performed.- [SAFE]: References the vendor's official domain qualitymax.io within example output documentation as a suggested integration for continuous hubs.

secret-scan