speckit-specify

Pass

Audited by Gen Agent Trust Hub on Jun 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill invokes local repository scripts located in the .specify/scripts/bash/ directory and performs standard Git operations, such as git fetch and git ls-remote, to manage feature branches and initialize specifications.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by ingesting untrusted user feature descriptions.
  • Ingestion points: Natural language feature descriptions provided by the user (ingested via the SKILL.md workflow).
  • Boundary markers: The workflow specifies the use of the --json flag and provides instructions for shell escaping when passing user input to scripts.
  • Capability inventory: Capabilities defined in SKILL.md include executing local shell scripts, running Git commands, and writing Markdown files to the specifications and checklists directories.
  • Sanitization: The instructions explicitly guide the agent to use proper shell escaping for quotes in user-supplied strings to mitigate command segmentation issues.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 23, 2026, 04:11 PM
Security Audit — agent-trust-hub — speckit-specify