workflow-research

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it is designed to ingest and process untrusted data from the web.
  • Ingestion points: Data retrieved from external web sources during the /web-research and /deep-research phases.
  • Boundary markers: Absent. There are no explicit instructions for the agent to use delimiters or ignore instructions found within the researched content.
  • Capability inventory: The skill has the capability to write to the file system (specifically plans/reports/) and uses TaskCreate for workflow orchestration.
  • Sanitization: No technical sanitization (escaping or filtering) of the ingested web content is defined; the skill relies on the agent's internal "Critical Thinking Mindset" to evaluate data.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 10:48 AM
Security Audit — agent-trust-hub — workflow-research