shrimp-task-manager

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements a robust, multi-stage task management workflow including planning, technical analysis, reflection, and verification.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it is designed to ingest and process external project documentation such as Product Requirements Documents (PRDs). Evidence: Ingestion points include the 'description', 'requirements', and 'summary' parameters in SKILL.md; boundary markers for separating user data from instructions are not defined; the skill has capabilities to read files and update project state; no sanitization logic is specified.
  • [COMMAND_EXECUTION]: The skill defines various tools for structured planning and execution guidance, all of which operate within the intended scope of project task management.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 05:31 AM
Security Audit — agent-trust-hub — shrimp-task-manager