speckit-analyze

Warn

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes a local shell script at '.specify/scripts/bash/check-prerequisites.sh' during the initialization phase to verify project state.
  • [COMMAND_EXECUTION]: Implements a dynamic hook system that reads command strings from the project's '.specify/extensions.yml' file and executes them using the 'EXECUTE_COMMAND' directive both before and after analysis.
  • [PROMPT_INJECTION]: Subject to indirect prompt injection vulnerabilities due to its design. 1. Ingestion points: Reads contents from 'spec.md', 'plan.md', 'tasks.md', and '.specify/extensions.yml'. 2. Boundary markers: No explicit delimiters or instructions are used to separate untrusted artifact content from the agent's internal logic. 3. Capability inventory: Includes the ability to execute shell scripts and dynamic commands based on file content. 4. Sanitization: While the skill includes basic escaping for shell arguments in the prerequisite script, it lacks validation or sanitization for the instructions and commands parsed from project configuration files.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 30, 2026, 09:09 PM
Security Audit — agent-trust-hub — speckit-analyze