# Check tool results, pages and uploads for hidden instructions before the agent reads them (slim)

> Check tool results, pages and uploads for hidden instructions before the agent reads them: 0 companies fully do it, 12 with a limit, 16 carry the tag. Every listing in Guardrails & safety filters with the reason and a source.

- Full: https://www.anchorterminal.com/niches/guardrails.screen-untrusted-content/index.md (~1,700 tokens) · this version ~430 tokens · JSON https://www.anchorterminal.com/niches/guardrails.screen-untrusted-content/index.json · canonical https://www.anchorterminal.com/niches/guardrails.screen-untrusted-content/
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · MCP: https://www.anchorterminal.com/mcp · Updated: 2026-11-15
- Licence: CC BY 4.0, credit Anchor Terminal and link this page (https://www.anchorterminal.com/terms/#data)

No company's docs show a full fit yet. The listings below do the job with a limit, or not at all. Full fit: 0 companies, with a limit: 12, tagged: 16. Weight 16, 9 accounts.

Done when: The content is submitted in a field marked as untrusted, separate from the user prompt, attacks phrased as harmless requests are scored as well as 'ignore previous instructions', the agent's own framing text is not flagged, and text longer than the input limit is split with overlap.
caveat: agentgate (Mohit Singh (individual))
caveat: amazon-bedrock-guardrails (Amazon Web Services, Inc.)
caveat: azure-ai-content-safety (Microsoft Corporation)
caveat: cisco-ai-defense-inspection (Cisco Systems, Inc.)
caveat: google-model-armor (Google LLC)
caveat: guardrails-ai (Guardrails AI, Inc.)
caveat: lakera-guard (Check Point Software Technologies Ltd.)
caveat: llamafirewall (Meta Platforms, Inc.)
caveat: mistral-moderation (Mistral AI (RCS Paris 952 418 325))
caveat: nemo-guardrails (NVIDIA Corporation)
caveat: openai-guardrails (OpenAI (as named in the LICENSE copyright line and the PyPI author field))
caveat: prisma-airs (Palo Alto Networks, Inc.)

Data: https://www.anchorterminal.com/api/v1/use-cases/guardrails.screen-untrusted-content.json
