clj-discover

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses shell commands to interact with the brepl tool for Clojure code execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents an attack surface for indirect prompt injection by processing and expanding macros from the project environment.
  • Ingestion points: The skill reads macro names and arguments from the code being analyzed in the workspace.
  • Boundary markers: No specific boundary markers or safety instructions are provided to delimit untrusted macro code from the agent's own instructions.
  • Capability inventory: The brepl tool allows execution of Clojure code, which has the capability to perform network and file system operations.
  • Sanitization: The skill directly interpolates macro code into the macroexpand command without validation.
  • [DYNAMIC_EXECUTION]: The skill generates and executes Clojure code strings at runtime to reflect on Java classes and analyze macros.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:37 AM
Security Audit — agent-trust-hub — clj-discover