The Agent Skills Directory

[SAFE]: The skill functions as a documentation and methodology guide for security professionals. All recommended actions and tools align with legitimate security research and auditing workflows. No malicious behaviors such as credential theft or unauthorized persistence were found within the instructions or provided files.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external project data, creating a potential surface for indirect prompt injection from the target codebase being analyzed.
Ingestion points: Target project files including documentation, configuration manifests (e.g., package.json, requirements.txt), and source code files accessed via shell commands like grep, cat, and find in Phase 1 through Phase 5.
Boundary markers: The methodology does not explicitly instruct the agent to use boundary markers or delimiters when reading and processing untrusted project content.
Capability inventory: The skill leverages various local capabilities including shell execution (to run commands like tree, grep, and ls) and security-specific CLI tools (semgrep, bandit, trufflehog).
Sanitization: There is no explicit instruction to sanitize or escape data read from the target project before processing it in the agent context.

zz-code-recon