add-archon-model
Pass
Audited by Gen Agent Trust Hub on May 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides legitimate instructional content for software development and does not contain any malicious code, obfuscation, or credential theft patterns. All external references are to well-known services like HuggingFace.
- [PROMPT_INJECTION]: The skill's primary workflow involves an indirect prompt injection attack surface because it requires the agent to ingest and analyze untrusted external data.
- Ingestion points: Step 1 instructions (SKILL.md) direct the agent to fetch and read
config.jsonandmodeling_*.pyfiles from HuggingFace repositories. - Boundary markers: Absent. The skill does not recommend using specific delimiters or safety warnings when the agent processes the external code.
- Capability inventory: The agent possesses file-writing capabilities (creating
model.py,spec.py, etc.), network access (fetching from HuggingFace), and potential command execution (runningpytestfor verification). - Sanitization: Absent. There is no explicit requirement to sanitize or validate the external code before it is used to influence the agent's code generation logic.
Audit Metadata