skill-compatibility-checker
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The script uses
subprocess.runto callopenclaw --version. This is a hardcoded, safe command used for system version detection. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external
SKILL.mdfiles to check for version constraints. Ingestion points: Reads frontmatter from skills located in directories under~/.openclaw/extensions/superpowers/skills/viacheck.py. Boundary markers: Explicitly searches for YAML frontmatter delimiters (---) and parses only that section. Capability inventory: Limited to reading local files, writing a state file, and executing a static version-check command. Sanitization: Employsyaml.safe_load()to parse untrusted metadata safely, preventing unsafe object instantiation.
Audit Metadata