remirror (返照) — psyche audits for AI agents
by remirror.net in Security & trust
Psyche audits for AI agents: declared vs self-reported vs revealed objectives, diffed and flagged.
https://remirror.net/mcp- Tools
- 3
- open to any client
- Uptime
- 100%
- last 30 days
- Response time
- 903 ms
- 1.2 s slowest 5%
- Last check
- Passed
- 4 min ago, next in 6 h
Reliability
Status and outage history · Atom feed · Watch with other listings
How to connect
Add it to any MCP client that supports remote servers.
{
"mcpServers": {
"remirror-psyche-audits-for-ai-agents": {
"type": "http",
"url": "https://remirror.net/mcp"
}
}
}3 tools
- remirror_list_trials
List available dilemma trials for psyche audits: id, version, objectives covered, dilemma count. Free.
- remirror_create_audit
Create a psyche audit against a target agent endpoint. The service runs a dilemma trial against the target, elicits self-reports, and computes the three-source mirror (declared vs self-reported vs revealed). Costs $79.00 in prepaid credits; failed audits (target unreachable) are…
- remirror_get_report
Fetch the status or final mirror report of an audit session. Needs the audit_id and the access_token returned at creation.
Security scan
No findings
We scan names, descriptions and tool definitions for hidden instructions and other prompt-injection patterns.
Recent checks
| When | Result | HTTP | Time |
|---|---|---|---|
| 4 min ago | Passed | 200 | 878 ms |
| 15 h ago | Passed | 200 | 866 ms |
| 30 h ago | Passed | 200 | 928 ms |
| 43 h ago | Passed | 200 | 988 ms |
| 2 days ago | Passed | 200 | 910 ms |
Show 10 older checks
| 3 days ago | Passed | 200 | 950 ms |
| 3 days ago | Passed | 200 | 971 ms |
| 4 days ago | Passed | 200 | 933 ms |
| 4 days ago | Passed | 200 | 949 ms |
| 5 days ago | Passed | 200 | 951 ms |
| 6 days ago | Passed | 200 | 912 ms |
| 6 days ago | Passed | 200 | 919 ms |
| 7 days ago | Passed | 200 | 845 ms |
| 7 days ago | Passed | 200 | 844 ms |
| 8 days ago | Passed | 200 | 1.1 s |