spec-implementation-plan

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a PowerShell block to identify the repository root and source a local environment configuration script (spec-common.ps1). It also instructs the agent to generate specific PowerShell commands for implementation tasks within the plan.md output.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from requirement and design documents to generate implementation plans that contain executable tasks.
  • Ingestion points: Reads data from {FEATURE_DIR}/requirements/solution.md, {FEATURE_DIR}/requirements/prd.md, and {FEATURE_DIR}/design/design.md.
  • Boundary markers: The skill does not use delimiters or safety instructions to wrap or isolate content interpolated from input files into the plan template.
  • Capability inventory: The skill writes to the local file system (plan.md) and generates shell commands intended for execution in subsequent workflow stages.
  • Sanitization: No validation, escaping, or filtering of the extracted text from requirement or design documents is performed prior to inclusion in the generated implementation plan.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 07:48 AM