form-1099-sa
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external data from tax documents (Form 1099-SA) and user-supplied medical expense descriptions, which creates a potential vector for indirect prompt injection.
- Ingestion points: The agent is instructed in
SKILL.md(Workflow Steps 3 and 4) to read Box values from Form 1099-SA and process user-supplied text describing medical expenses and receipt statuses. - Boundary markers: Absent. There are no instructions to the agent to employ delimiters or explicitly ignore embedded instructions within the processed data.
- Capability inventory: The skill includes a detailed playbook in
filing.mdfor using browser automation tools (Playwright, Puppeteer, Selenium, or Browserbase) to interact with IRS websites and perform filing operations. It also produces file-based drafts for the user. - Sanitization: Absent. The skill does not define specific validation, escaping, or filtering logic for the data before it is interpolated into filing flows or draft outputs.
Audit Metadata