cowork-rfp-response

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill ingest data from external RFP documents and internal company material folders, making it susceptible to indirect prompt injection attacks.
  • Ingestion points: The skill processes RFP documents and various company materials (past proposals, case studies, bios, certifications) as specified in the workflow.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the document content as raw data or to ignore instructions embedded within the files.
  • Capability inventory: The skill utilizes powerful tools including Read, Glob, Grep, Write, Bash, and WebSearch, which could be misused if an injection is successful.
  • Sanitization: There is no evidence of sanitization or validation of the text extracted from the source documents before it is processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 06:05 PM
Security Audit — agent-trust-hub — cowork-rfp-response