pstack-statechart

Pass

Audited by Gen Agent Trust Hub on Oct 24, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill downloads the TLC model checker (tla2tools.jar) from the official TLA+ GitHub repository and Java (Temurin) via the mise tool manager. These downloads are performed with pinned versions and include a SHA-256 checksum for the jar file to ensure integrity.
  • [COMMAND_EXECUTION]: The skill executes npm install to fetch rendering libraries (pele and @resvg/resvg-js) and uses mise to install and manage the TLA+ execution environment. These commands are hardcoded, targeted at specific tool management tasks, and utilize security features like restricted path lookups (MISE_CEILING_PATHS).
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied TLA+ specifications and Mermaid chart files to generate its output. While this ingestion of external data theoretically exposes an attack surface for indirect prompt injection, the skill operates in a restricted environment without access to sensitive credentials or persistent system modifications.
  • [SAFE]: The skill follows security best practices for external resource management, including the use of temporary directories for processing and explicit verification of downloaded assets.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 24, 2026, 04:52 PM