practice-cognition
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill implements a structured problem-solving methodology ("Practice-cognition loop") that guides the agent through sensory perception, rational hypothesis formation, and practice-based verification. It does not contain any malicious patterns or instructions.
- [NO_CODE]: The skill consists entirely of Markdown instructions and philosophical citations. It does not include any scripts, binary files, package dependency files (e.g., requirements.txt, package.json), or remote code execution patterns.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a process for ingesting external data during the "Sensory perception" phase (reading code, running programs, observing output). While this is a standard ingestion point for untrusted data, the skill's structured output template (Hypothesis -> Expected -> Observed -> Evaluation) encourages the agent to treat data as evidence for validation rather than authoritative instructions.
Audit Metadata