case-study-prep

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, which creates a potential surface for indirect prompt injection attacks.
  • Ingestion points: Untrusted data is retrieved from user-provided Google Docs via the google_drive_fetch tool and from Slack thread responses generated by the eon bot.
  • Boundary markers: The instructions do not specify the use of delimiters or protective markers to separate the ingested customer notes from the agent's core processing logic.
  • Capability inventory: The skill possesses the capability to write files to the system (generating .docx files) and interact with Slack messaging.
  • Sanitization: There is no evidence of content sanitization or instruction-filtering for the data fetched from external sources before it is synthesized into the final document.
  • [COMMAND_EXECUTION]: The skill references a validate.py script and a docx-js tool. These appear to be internal platform capabilities or associated skills used for standard document validation and generation within the expected workflow.
  • [DATA_EXFILTRATION]: The skill reads data from Slack and Google Drive to produce a local document. No network exfiltration to non-whitelisted third-party domains was detected; interactions are limited to the platform's internal tools and the Slack bot 'eon'.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 01:21 PM