skill-ecosystem-doctor

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/ecosystem_scans.py uses subprocess.run to call the loom binary for workspace diagnostics. The call uses an argument list rather than a raw shell string, which is a secure method for invoking local tools.
  • [DATA_EXFILTRATION]: The skill implements a scanner in scripts/ecosystem_scans.py that searches for high-confidence secret patterns like API keys and private keys. The skill's design ensures that these values are suppressed in reports and treated only as security findings to be remediated.
  • [PROMPT_INJECTION]: The skill processes numerous external SKILL.md files and governance documents, creating a surface for indirect prompt injection. A malicious skill in the scanned ecosystem could attempt to influence the agent's behavior. 1. Ingestion points: Scans the filesystem for SKILL.md files and governance JSON in roots defined by the source policy. 2. Boundary markers: The skill does not currently use specific delimiters or ignore-embedded-instructions warnings for audited content. 3. Capability inventory: Includes file system modification (linking, unlinking, writing), configuration management, and execution of diagnostics binaries. 4. Sanitization: Employs safe parsing for structured data (SafeLoader for YAML) and provides redaction logic for detected secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 12:36 AM
Security Audit — agent-trust-hub — skill-ecosystem-doctor