happycapy-skill-creator

Pass

Audited by Gen Agent Trust Hub on Mar 21, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Clones repository data from the anthropics/skills GitHub repository to provide a base for skill generation.
  • [COMMAND_EXECUTION]: Uses subprocess.run to execute git clone and chmod for managing repository content and setting script permissions.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it incorporates untrusted data from user requirements and external source code into LLM prompts to create new Python scripts. Malicious content within the source data could potentially steer the LLM to generate harmful code.
  • Ingestion points: User-provided requirement strings and file content from the anthropics/skills repository.
  • Boundary markers: Lacks explicit sanitization or instructions to ignore instructions embedded within the source data.
  • Capability inventory: File writing, permission management, and Git command execution.
  • Sanitization: No content filtering or validation is performed on the data processed by the LLM.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 21, 2026, 07:18 AM