dsh-doc-standards

Pass

Audited by Gen Agent Trust Hub on Aug 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONNO_CODE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilize standard development tools including git, grep, xargs, and wc, along with project-specific scripts via pnpm run (e.g., verify-doc-budgets, doc-sync). These tools are used solely for static analysis, linting, and word count reporting on local repository files.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill analyzes repository documentation and source code to verify standards. It does not attempt to access sensitive files such as credentials, SSH keys, or environment variables, and it performs no network operations to external domains.
  • [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection as it ingests Markdown files and code comments. 1. Ingestion points: Markdown documents (.md) and TypeScript source files in the deepseek-harness repository. 2. Boundary markers: None explicitly mentioned in the skill. 3. Capability inventory: Local file read/write, git, grep, and pnpm script execution. 4. Sanitization: None. This represents an attack surface for processing untrusted documentation but lacks high-privilege triggers.
  • [SAFE]: No indicators of obfuscation, persistence mechanisms, privilege escalation, or remote code execution from unknown sources were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 16, 2026, 04:08 PM
Security Audit — agent-trust-hub — dsh-doc-standards