speckit-implement-zh

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes multiple repository files to determine its implementation steps, creating a surface for indirect instructions to influence agent behavior.
  • Ingestion points: Reads and analyzes tasks.md, plan.md, research.md, and various markdown files in the checklists/ and contracts/ directories.
  • Boundary markers: The instructions do not define specific delimiters or use "ignore embedded instructions" warnings when processing the contents of these files.
  • Capability inventory: The skill has the capability to write code, modify files, and execute scripts based on the plan extracted from these documents.
  • Sanitization: The skill does not perform explicit validation or sanitization of the markdown content before using it to guide its implementation logic.- [COMMAND_EXECUTION]: The skill executes local system commands and scripts to identify the project environment and configuration.
  • Evidence: Usage of git rev-parse to detect Git repositories and invocation of PowerShell/Bash scripts located in the .specify/scripts/ directory for prerequisite checks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 06:04 PM
Security Audit — agent-trust-hub — speckit-implement-zh