recheck-against-upstream-deep

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill is designed to ingest and process external documentation, which is a potential surface for indirect prompt injection if the source documentation contains instructions intended to influence the agent. 
  • Ingestion points: The skill instructs the agent to fetch "CURRENT official upstream docs" for various surfaces (config, APIs, etc.) in SKILL.md. 
  • Boundary markers: The instructions do not specify unique delimiters or safety prompts for the subagents when parsing external content. 
  • Capability inventory: The skill utilizes session-based memory for checkpointing and can interface with external work-item / issue-tracker capabilities as described in SKILL.md. 
  • Sanitization: The skill relies on the agent's interpretation of fetched text without explicit sanitization steps mentioned in the instructions.
  • [SAFE]: The skill implements throttling and checkpointing as best-practice reliability measures for heavy tasks and correctly routes single-surface requests to simpler sibling tools to minimize resource usage.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 07:56 PM
Security Audit — agent-trust-hub — recheck-against-upstream-deep