spec-kit-vibe-compat

Warn

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The usage instructions in SKILL.md suggest executing a PowerShell script with the -ExecutionPolicy Bypass flag. This flag instructs PowerShell to ignore local execution policies that may be configured to prevent the running of unsigned or untrusted scripts.
  • [PROMPT_INJECTION]: The scripts/speckit-router.ps1 script interpolates user-provided arguments directly into a string output presented as a recommended prompt for the agent, creating a surface for indirect prompt injection.
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface Evidence: 1. Ingestion points: Command line arguments ($ArgsIn) in scripts/speckit-router.ps1. 2. Boundary markers: None; user input is directly concatenated to the command string. 3. Capability inventory: Local file reads (command-map.json) and terminal output (Write-Host). 4. Sanitization: No input validation or escaping is applied to the user-supplied string before interpolation.
Audit Metadata
Risk Level
MEDIUM
Analyzed
May 11, 2026, 02:49 PM
Security Audit — agent-trust-hub — spec-kit-vibe-compat