tlaplus
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill includes instructions to download the
tla2tools.jarmodel checking utility from the official TLA+ GitHub repository. This is a trusted and well-known source for formal verification software. - [COMMAND_EXECUTION]: The skill guides users to execute shell commands like
tlcandjava -jar tla2tools.jarto run the model checker on local TLA+ specifications. These commands are standard for this domain and are used within their intended scope. - [INDIRECT_PROMPT_INJECTION]: The skill ingests system design descriptions from the user to generate TLA+ code. While this creates a surface for indirect instructions, the risk is mitigated by the specific context of formal modeling and the limited capabilities of the verification tools.
- [DYNAMIC_EXECUTION]: User-provided or agent-generated TLA+ specifications are executed by the TLC model checker as part of the verification process. This dynamic execution is essential for discovering design flaws like race conditions and is performed using established tools locally.
Audit Metadata