task

Pass

Audited by Gen Agent Trust Hub on Mar 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill implements a standard development workflow, utilizing project-local context and standard file operations to assist in task planning and documentation.
  • [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it aggregates data from the codebase and user requests into task documents that guide subsequent agent actions.
  • Ingestion points: User input and file content retrieved through Glob and Grep operations (SKILL.md).
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to prevent the agent from obeying instructions embedded within the codebase content it embeds into task documents.
  • Capability inventory: The skill can create and modify files (docs/task/*.md, TASKS.md) and trigger the execution of sub-agents using the Task() orchestration command (SKILL.md).
  • Sanitization: There is no mention of sanitization or validation of the codebase content before it is interpolated into the planning documents.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 26, 2026, 06:04 AM
Security Audit — agent-trust-hub — task