Fact-check and fix AI outputs. Catches hallucinations, repairs broken JSON, corrects errors — before they reach users.
Fact-check and fix AI outputs. Catches hallucinations, repairs broken JSON, corrects errors — before they reach users. Exposed over MCP by the perf mcp mcp server, that capability becomes something an assistant can invoke while it works, not something you go and do afterwards.
Setup follows the usual MCP pattern — install or clone the server, register it in your client's configuration file, restart the client. The configuration blocks on this page cover the common clients.
Everything the assistant can do here goes through one of these:
perf_verify — Detect and repair hallucinations in LLM-generated text. Uses multi-channel verification (web search, NLI models, cross-reference) — not just anotherperf_validate — Validate LLM-generated JSON against a schema and auto-repair violations. Fixes malformed enums, wrong types, missing fields, hallucinated propertiesperf_correct — General-purpose output correction. Classifies the error type and applies the right fix — hallucination, schema violation, semantic inconsistency, orperf_chat — Route LLM requests to the optimal model automatically. Selects between GPT-4o, Claude, Gemini, and 20+ models based on task complexityCursor — The Cursor tool exposed by this serverWindsurf — The Windsurf tool exposed by this serverYou will need 2 environment variables: PERF_API_KEY, PERF_BASE_URL. The server will not start without them, which is usually why the tools fail to appear on a first run. Keep credentials in your client's env block or a secrets manager rather than in a file you might commit.
This sits in the AI and media services group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Perf MCP's toolset — perf_verify, perf_validate, perf_correct and 3 more — is a fair guide to whether it matches your workflow. It is maintained by perf-technology; worth a glance at recent repository activity before you build anything load-bearing on it.
We check each listing at SyncDev against the project's documentation before it goes live — if something here drifts out of date, it is a bug worth reporting.
| Tool | What it does |
|---|---|
| perf_verify | Detect and repair hallucinations in LLM-generated text. Uses multi-channel verification (web search, NLI models, cross-reference) — not just another LLM check. |
| perf_validate | Validate LLM-generated JSON against a schema and auto-repair violations. Fixes malformed enums, wrong types, missing fields, hallucinated properties. |
| perf_correct | General-purpose output correction. Classifies the error type and applies the right fix — hallucination, schema violation, semantic inconsistency, or instruction drift. |
| perf_chat | Route LLM requests to the optimal model automatically. Selects between GPT-4o, Claude, Gemini, and 20+ models based on task complexity. OpenAI-compatible format. |
| Cursor | The Cursor tool exposed by this server. |
| Windsurf | The Windsurf tool exposed by this server. |
{
"mcpServers": {
"perf": {
"command": "npx",
"args": ["-y", "perf-mcp"],
"env": {
"PERF_API_KEY": "pk_live_xxx"
}
}
}
}Configuration as documented by the project. Restart the client after saving.
| Variable | Description | Required |
|---|---|---|
| PERF_API_KEY | Credential the server authenticates with. | Yes |
| PERF_BASE_URL | Endpoint or connection string the server talks to. | Yes |
Build a programmable telecommunications stack for connecting telephony services with the Internet via a cloud-based utility.
Search built for AI, not humans — semantic web search that returns model-ready content, plus code context.
Answers, not links — delegate questions to Perplexity's search-grounded models and get cited responses back.
Give your assistant a voice — text-to-speech, voice cloning and audio tools from the ElevenLabs API.
Give your assistant a real code sandbox — isolated cloud VMs for actually running the code it writes.
The ML hub in your context window — search models, datasets, papers and run Spaces from the official server.