maintain-formal-verification-skill
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to identify and run shell commands found within project-specific scripts or 'verify skills' to execute model checkers, provers, or other code-level verifiers.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from the repository, including commit messages, source code diffs, documentation comments, and changelogs, to decide how to update verification models and properties.
- Ingestion points: Repository content including source files, commit logs, and documentation accessed during Step 3 (Read every change to covered code).
- Boundary markers: The instructions do not specify the use of delimiters or 'ignore' instructions when the agent processes these external data sources.
- Capability inventory: The skill possesses the ability to execute shell commands (Step 2 and 5) and write to local files to update verification models, property files, and skill instructions (Step 4 and 6).
- Sanitization: There is no mention of sanitizing or validating the ingested repository content before it is used to influence the agent's logic or the content of file modifications.
Audit Metadata