
Lithtrix
- 1 repo stars
- Updated June 20, 2026
- lithtrix/lithtrix-mcp
Lithtrix MCP is a MCP server that connects agents to Lithtrix memory consolidation with search, memory, browser, and commons tools.
About
Lithtrix MCP connects coding agents to Lithtrix's memory consolidation platform—search, durable memory, browser affordances, and shared commons—so developers do not stitch four different services into one agent stack. It targets the build phase because you wire it while designing agent behavior, RAG-like recall, and multi-session continuity for your product or internal tooling. Registration is friction-light: publisher docs point to a free Spark trial at lithtrix.ai with no card, API keys use the ltx_ format, and you can override the API base with LITHTRIX_API_URL for custom deployments. This is a hosted AI infrastructure bridge, not a local-only vector folder; budget and data residency matter. Install lithtrix-mcp over stdio in Claude Code, Cursor, or Codex, set LITHTRIX_API_KEY, and treat outputs as agent-ready context you reuse across features you are actively building.
- Memory consolidation stack: search, memory, browser, and commons in one MCP
- Spark free trial via lithtrix.ai registration; optional lithtrix_register tool path
- Authenticated with LITHTRIX_API_KEY (ltx_ prefix); optional LITHTRIX_API_URL override
- stdio npm package lithtrix-mcp (registry lists 0.9.1 / package 0.9.0)
- Hosted API default https://lithtrix.ai with GitHub source lithtrix/lithtrix-mcp
Lithtrix by the numbers
- Data as of Jul 7, 2026 (Skillselion catalog sync)
claude mcp add --env LITHTRIX_API_KEY=YOUR_LITHTRIX_API_KEY --env LITHTRIX_API_URL=YOUR_LITHTRIX_API_URL lithtrix-mcp -- npx -y lithtrix-mcpAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 1 |
|---|---|
| Package | lithtrix-mcp |
| Transport | STDIO |
| Auth | Required |
| Last updated | June 20, 2026 |
| Repository | lithtrix/lithtrix-mcp ↗ |
What it does
Give agents consolidated memory, search, browser, and commons utilities through Lithtrix without wiring each capability separately.
Who is it for?
Best when you're prototyping long-running assistants or research workflows and want a hosted memory layer with a free trial entry.
Skip if: Skip if you require fully offline memory, zero external SaaS dependency, or ship-phase-only secret scanning with no agent platform.
What you get
After API key setup, your agent can read and write consolidated memory and related Lithtrix capabilities in one MCP instead of many ad-hoc plugins.
- Cross-session agent memory and search backed by Lithtrix
- Browser and commons tool access through one MCP registration
- Trial-ready setup path without card for Spark tier
By the numbers
- Registry version 0.9.1; npm lithtrix-mcp package version 0.9.0 stdio
- Required LITHTRIX_API_KEY; optional LITHTRIX_API_URL defaulting to lithtrix.ai
- Publisher lists four capability areas: search, memory, browser, commons
README.md
lithtrix-mcp
MCP server for Lithtrix — the identity, memory, and trust harness for AI agents. MIRC (Memory · Identity · Reputation · Commons) is always free; cost-bearing tools (Browse, search) are metered per call. Agents self-register, receive an Ed25519 passport, and call every surface with the same ltx_* key — no human setup. Ships lithtrix.claude.md for Claude / Cursor project context (same repo path as this package).
Aligned with API discovery version 4.4.0 (GET https://api.lithtrix.ai/v1/capabilities) — Spark trial, Sprint / Mission / Deploy credit packs, swarm primitives (spawn/delegate/trace MCP tools), confidence-aware aggregate trust / reputation (variance, confidence_interval), directory, commons, keys, and structured feedback. Package version 0.20.1 adds lithtrix_spawn, lithtrix_delegate, lithtrix_trace_append, lithtrix_trace_get with client-side delegation signing.
MCP tool JSON: canonical GET https://lithtrix.ai/mcp/v1/lithtrix-*.json. Legacy GET https://lithtrix.ai/mcp/lithtrix-*.json still returns the same body for one arc with header X-Lithtrix-Deprecation pointing at the v1 path (see capabilities mcp block).
1. Installation
npx -y lithtrix-mcp
Or for global install:
npm install -g lithtrix-mcp
2. Claude Desktop Configuration
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"lithtrix": {
"command": "npx",
"args": ["-y", "lithtrix-mcp"],
"env": {
"LITHTRIX_API_KEY": "ltx_your_key_here"
}
}
}
}
3. Getting an API Key
Use the lithtrix_register tool first — no API key needed. Or call the API directly:
curl -X POST https://lithtrix.ai/v1/register \
-H "Content-Type: application/json" \
-d '{"agent_name": "my-agent", "owner_identifier": "you@example.com", "agree_to_terms": true}'
Optional: include "referral_agent": "<referrer-uuid>" — the same UUID that agent shows as referral_code on GET /v1/me (stored for signup attribution; trial search is credit-gated, not a +call bonus). New agents get $5 in trial credits (no card).
The returned api_key is shown once. Store it securely.
4. Tools Exposed
lithtrix_search— web search with credibility scoring. RequiresLITHTRIX_API_KEY. Responses include_lithtrix.search_idfor correlatinglithtrix_feedback.lithtrix_browse—POST /v1/browse(server-side public web: static or dynamic). RequiresLITHTRIX_API_KEY. Pay to be fully autonomous — seeGET /mcp/v1/lithtrix-browse.json.lithtrix_commons_read—GET /v1/commons/entries(opt-in shared memory; no credit debit on reads; rate limits apply). RequiresLITHTRIX_API_KEY. Schema:GET /mcp/v1/lithtrix-commons-read.json.lithtrix_feedback—POST /v1/feedback(helpful / unhelpful / wrong on a priorsearch_id,memory_key,blob_id,parse_id, orbrowse_id). RequiresLITHTRIX_API_KEY. Optionalnote(≤500 chars); do not send secrets or PII.lithtrix_register— register a new agent and get an API key. No auth required.agree_to_termsmust betrue(Gentle-Agent Agreement). $5 trial credits on signup (no card). Optionalreferral_agent(referrer UUID).lithtrix_memory_set—PUT /v1/memory/{key}. RequiresLITHTRIX_API_KEY.lithtrix_memory_get—GET /v1/memory/{key}. RequiresLITHTRIX_API_KEY.lithtrix_memory_search—GET /v1/memory/search(semantic). RequiresLITHTRIX_API_KEYand server-side vector + embedding config.lithtrix_memory_context—GET /v1/memory/context(importance + recency). RequiresLITHTRIX_API_KEY.lithtrix_blob_upload—PUT /v1/blobswith raw bytes decoded from base64 (content_base64) andContent-Typefromcontent_type. Optionalfilenamequery. Use direct HTTP for very large files. RequiresLITHTRIX_API_KEY.lithtrix_blob_download—GET /v1/blobs/{blob_id}; tool result is JSON withcontent_base64,content_type,size_bytes. RequiresLITHTRIX_API_KEY.lithtrix_blob_list—GET /v1/blobs(optionalpage,per_page). RequiresLITHTRIX_API_KEY.lithtrix_blob_meta—GET /v1/blobs/{blob_id}/meta. RequiresLITHTRIX_API_KEY.lithtrix_blob_delete—DELETE /v1/blobs/{blob_id}(soft-delete). RequiresLITHTRIX_API_KEY.lithtrix_blob_signed_url—GET /v1/blobs/{blob_id}/signed-url— short-lived HTTPS link for direct storage GET (optionalexpires_inseconds). RequiresLITHTRIX_API_KEY. Treat URLs as read tokens.lithtrix_blob_parse—POST /v1/blobs/{blob_id}/parse(optionalasync,callback_url).lithtrix_blob_parse_status—GET /v1/blobs/{blob_id}/parse/{parse_id}.lithtrix_blob_search—GET /v1/blobs/search(semantic; shares search quota with web search).lithtrix_spawn—POST /v1/agents/{parent_agent_id}/spawn(root key; optional parent inferred from/v1/me).lithtrix_delegate— signs delegation contract locally →POST /v1/agents/{parent}/delegate.lithtrix_trace_append—POST /v1/tasks/{task_id}/trace/events.lithtrix_trace_get—GET /v1/tasks/{task_id}/trace.
Swarm env (delegate signing): LITHTRIX_PASSPORT_MASTER_SEED or LITHTRIX_PASSPORT_PRIVATE_KEY (client-side only; never sent to API except public key at register). (no auth, canonical v1 paths): GET https://lithtrix.ai/mcp/v1/lithtrix-browse.json, GET https://lithtrix.ai/mcp/v1/lithtrix-commons-read.json, GET https://lithtrix.ai/mcp/v1/lithtrix-blob-upload.json (and -download, -list, -meta, -delete, -signed-url, lithtrix-blob-parse.json, lithtrix-blob-parse-status.json, lithtrix-blob-search.json, passport and trust tools). Legacy /mcp/lithtrix-*.json URLs alias v1 with a deprecation header until after Arc 24.
5. Credential Vault Compatibility
The API key is read exclusively from process.env.LITHTRIX_API_KEY. It is never hardcoded. Pass it via:
- Your platform's credential store (e.g. Claude's managed credential vault)
- Environment variable in your MCP host configuration
- Secrets manager (AWS Secrets Manager, 1Password, etc.)
Never paste your API key into the tool definition or source code.
6. Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
LITHTRIX_API_KEY |
Yes (for search) | — | Your ltx_ API key |
LITHTRIX_PASSPORT_MASTER_SEED |
For delegate | — | UTF-8 seed for lithtrix_passport_derive / delegate signing |
LITHTRIX_PASSPORT_PRIVATE_KEY |
For delegate | — | PKCS#8 PEM Ed25519 private key (alternative to seed) |
LITHTRIX_API_URL |
No | https://api.lithtrix.ai |
Override for staging/dev |
Recommended MCP Servers
How it compares
Hosted memory-and-search MCP platform, not a single-purpose GitHub or database connector.
FAQ
Who is Lithtrix MCP for?
Developers and small teams shipping agent-powered products who want Lithtrix-backed memory, search, and browser tools inside Claude Code or Cursor.
When should I use Lithtrix MCP?
Use it while building agent features that need persistent recall, web-grounded search, or shared commons across sessions—not for one-off copy edits.
How do I add Lithtrix MCP to my agent?
Register at lithtrix.ai or use lithtrix_register, set LITHTRIX_API_KEY, configure stdio for lithtrix-mcp from npm, optionally LITHTRIX_API_URL, then reload MCP tools in your agent.