speckit-implement

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local script at .specify/scripts/bash/check-prerequisites.sh and uses standard tools like git to verify project state and manage repository-specific ignore files.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because its core logic and execution flow are determined by the contents of files like tasks.md and plan.md.
  • Ingestion points: The skill reads and parses multiple project files, including tasks.md, plan.md, data-model.md, and research.md.
  • Boundary markers: There are no explicit instructions or delimiters used to separate user-provided task descriptions from the skill's operational instructions, which could allow malicious content in these files to influence agent behavior.
  • Capability inventory: The skill can create or modify files (ignore files) and execute arbitrary implementation tasks defined in the project files.
  • Sanitization: No sanitization or validation of the ingested file content is performed before it is integrated into the agent's prompt context.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 06:47 PM