docs-code-sample-standards
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides a comprehensive security and quality auditing framework for documentation code samples. It emphasizes detecting secret leaks, avoiding insecure defaults (like disabled TLS), and guarding destructive operations.
- [COMMAND_EXECUTION]: The skill includes a Python script (scripts/sample-audit.py) that the agent is instructed to run locally to perform mechanical linting of markdown files. This script uses only standard Python libraries (argparse, json, os, re, sys) and performs static regex-based analysis. It does not execute the code snippets it finds or make network connections.
- [CREDENTIALS_UNSAFE]: Reference files and evaluation cases include sample API key strings (e.g., sk_live_... patterns). These are explicitly defined as negative examples of security defects or are used as regex patterns to enable the script to detect and flag real credential leaks in user documentation. The skill actively promotes secret scanning and credential rotation.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and audit external documentation files. It mitigates associated risks through a highly structured auditing methodology involving specific rubrics, execution tiers, and sanitization rules for placeholders.
Audit Metadata