gs-maintain-formal-verification-skill
Pass
Audited by Gen Agent Trust Hub on Nov 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill is instructed to run formal verification checks using commands found in project-local configuration files. This is a core functional requirement for the skill's operation.- [DYNAMIC_EXECUTION]: The agent discovers and executes arbitrary commands defined in other skill files found in directories such as .cursor/skills/ or .claude/skills/ to perform its maintenance tasks.- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes untrusted data from source code files, git diffs, and commit messages. Ingestion points: Reads source code, git logs, and commit history to identify changes. Boundary markers: The instructions do not specify any delimiters or warnings to ignore embedded instructions in the ingested data. Capability inventory: The skill can execute shell commands (verification tools) and write to files (updating models and .gitignore). Sanitization: There are no explicit steps defined for sanitizing or escaping the content read from the code or git history. The risk is considered inherent to the developer-centric nature of the skill and is mitigated by the scope of its intended use.
Audit Metadata