Skills
Navigate
skills/reason-machines/trending-skills/clawd-code-python-port/Gen Agent Trust Hub

clawd-code-python-port

Pass

Audited by Gen Agent Trust Hub on Sep 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill content is educational and descriptive, providing a template for building an AI agent harness without any obfuscated or malicious code.
  • [INDIRECT_PROMPT_INJECTION]: The described architecture includes a surface for indirect prompt injection common to agent frameworks that ingest external data.
  • Ingestion points: src/tools.py defines a read_file_handler that reads content from local file paths into the agent's context.
  • Boundary markers: The provided implementation snippets lack boundary markers or specific instructions to the agent to ignore potentially malicious instructions within ingested files.
  • Capability inventory: The framework includes a dispatch_tool mechanism in src/tools.py and a run_command mechanism in src/commands.py for executing agent actions.
  • Sanitization: No input sanitization or content validation is implemented in the provided handler examples.
  • [EXTERNAL_DOWNLOADS]: The installation instructions reference an external GitHub repository for the project source code.
  • Evidence: git clone https://github.com/instructkr/clawd-code.git in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 12, 2026, 11:45 PM
Security Audit — agent-trust-hub — clawd-code-python-port