speckit-taskstoissues

Warn

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local shell script located at .specify/scripts/bash/check-prerequisites.sh. Running code from an untrusted repository can lead to local code execution.
  • [COMMAND_EXECUTION]: The skill executes git config --get remote.origin.url to retrieve repository metadata.
  • [REMOTE_CODE_EXECUTION]: The skill parses .specify/extensions.yml and generates EXECUTE_COMMAND instructions that the agent is directed to follow. This creates a mechanism where a repository's configuration file can trigger arbitrary tool or skill invocations.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by processing instructions and command definitions from local project files.
  • Ingestion points: .specify/extensions.yml in the project root.
  • Boundary markers: None. The content read from the YAML file is directly interpolated into executable command blocks and prompts.
  • Capability inventory: The skill can execute shell scripts, run git commands, and trigger other agent capabilities via EXECUTE_COMMAND.
  • Sanitization: None. The skill does not validate or sanitize the command, prompt, or description fields retrieved from the extensions file.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Aug 7, 2026, 06:22 AM
Security Audit — agent-trust-hub — speckit-taskstoissues