# Jaxon (jaxon.ai)

Checks each AI output against user-written rules and returns true, false or unknown, over MCP, REST or on-premises.

- Atlas card: https://inite.ai/en/atlas/jaxon.ai
- Site: https://jaxon.ai
- For: teams whose wrong AI answers are unacceptable and need each output checked against written rules before action
- Use 0/100 · alive 100/100

## Used for

- check AI answers against written policy rules
- validate a model still meets its standard after changes
- compare model versions before release

## Price

Not read yet.

## For developers

- API docs: https://jaxon.ai/developers
- MCP server: ai.jaxon/dsail (namespace; auth on 2026-10-17)
- Read 2026-10-17

## MCP server, measured daily

Server: https://agents.jaxon.ai/mcp, checked every day since 2026-10-19 (8 checks).

- Answered: 100% of daily checks
- Median first answer: 249 ms
- Sign-in: OAuth (the client signs the user in)
- Method: https://inite.ai/en/atlas/mcp/report

## Your data

- Certifications stated: none
- Training on your data: yes (https://jaxon.ai/privacy)
- Read 2026-10-17

---

Source: INITE Atlas, https://inite.ai/en/atlas/jaxon.ai. JSON: https://inite.ai/api/atlas/jaxon.ai. MCP: https://inite.ai/api/mcp (atlas_lookup).