The Agent Skills Directory

[DATA_EXFILTRATION]: The skill is designed to search for and identify sensitive file paths (e.g., .env, .ssh/id_rsa) and hardcoded credentials in the source code as part of a security audit. This is a diagnostic function intended for local reporting and does not involve external transmission of data.
[COMMAND_EXECUTION]: The skill instructs the agent to execute established security command-line utilities, including semgrep, npm audit, pip-audit, govulncheck, cargo audit, and trivy. These are used appropriately within the context of a vulnerability scan.
[EXTERNAL_DOWNLOADS]: The use of semgrep --config auto involves fetching security rules from Semgrep's official registry. This is a standard operation for the tool and targets a well-known service.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted codebase data during the audit process.
Ingestion points: All files within the directory where the scan is initiated (SKILL.md, Phase 1.1 and 1.2).
Boundary markers: Absent; the skill does not specify delimiters or markers to isolate the content of scanned files from the agent's primary instructions.
Capability inventory: Subprocess execution of various audit tools (semgrep, npm, pip, trivy, cargo) and file-writing capability to create reports in the docs/security-audit/ folder.
Sanitization: Absent; the skill does not explicitly provide instructions for escaping or validating the content extracted from files before it is included in the final markdown report.

vulnerability-scan