one-task-management

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection due to its core functionality of ingesting and acting upon external task data.
  • Ingestion points: Untrusted data enters the agent context through the get-task-details, get-document, and search-tasks tools as described in SKILL.md.
  • Boundary markers: The skill instructions do not specify any delimiters or explicit warnings for the agent to ignore instructions embedded within the task or document bodies it processes.
  • Capability inventory: The skill has broad mutation capabilities, including create-initiative, update-initiative, patch-document, report-bug, and add-task-comment (SKILL.md).
  • Sanitization: There is no evidence of input validation or content escaping mentioned in the operating rules before the agent interpolates retrieved data into subsequent tool calls.
  • [SAFE]: Analysis of the skill's instructions, configuration, and assets found no evidence of hardcoded credentials, malicious remote code execution, or unauthorized network operations. The assets/logo.svg is a standard vector image with no embedded scripts or payloads.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 07:47 AM
Security Audit — agent-trust-hub — one-task-management