Tessera MCP Server
Tessera's MCP server surfaces LLM cost optimization decisions directly in your AI editor. Install it in Claude Desktop, Cursor, or Cline, pass a TESSERA_API_KEY, and you get six tools: list workloads with their anchor costs, pull savings reports, check queued recommendations with expected lift, browse audit-immutable ledger entries, view quality snapshots with drift events, and approve a mechanic to move it from suggested to active. The read tools let agents diagnose spend, the single write tool lets them act on recommendations with an audit trail. It's designed for teams routing LLM calls through Tessera's proxy and need agents to surface what's drifting or where to switch providers without opening the dashboard.
- tools
- 6
- auth
- Secret needed
- transport
- STDIO
- env vars
- 2
- last updated
- May 27, 2026
Tools
6 toolsPublic tool metadata for what this MCP can expose to an agent.
tessera_list_workloadsList your mapped workloads with anchor cost + current m-stack.
List your mapped workloads with anchor cost + current m-stack.
No parameter schema in public metadata yet.
tessera_get_savings_reportAnchored spend + measured savings for a window.
Anchored spend + measured savings for a window.
No parameter schema in public metadata yet.
tessera_get_recommendation_queuePending Optimize-tab recommendations with expected lift + confidence.
Pending Optimize-tab recommendations with expected lift + confidence.
No parameter schema in public metadata yet.
tessera_get_ledger_entriesAudit-immutable Monthly Reading rows (provider call, mechanic stack applied, savings).
Audit-immutable Monthly Reading rows (provider call, mechanic stack applied, savings).
No parameter schema in public metadata yet.
tessera_get_quality_snapshotSLA floor + p50/p95 quality scores + drift events.
SLA floor + p50/p95 quality scores + drift events.
No parameter schema in public metadata yet.
tessera_approve_recommendationMove a queued mechanic from "suggested" to "active" with audit-trail entry.
Move a queued mechanic from "suggested" to "active" with audit-trail entry.
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
TESSERA_API_KEY*secretAPI key in tk_* format. Get one free at https://tesseraai.io/dev.
TESSERA_UPSTREAM_API_BASE_URLdefault: https://ledger.tesseraai.ioOverride only for staging or local-dev testing.
Readme
tessera-mcp-server
The MCP server that returns money, not data.
Status: v0.1.3 — published 2026-05-27 on npm with sigstore SLSA provenance v1, listed on registry.modelcontextprotocol.io.
Tessera is an LLM proxy that optimizes API spend through multi-provider routing, prompt compression, audit-immutable logging, output-length prediction, and batch arbitrage. This package exposes Tessera as an MCP server for tool-using agents — Claude Desktop, Claude Code, Cursor, Cline, Continue, Goose, Zed.
Where other LLM-infrastructure MCP servers return logs, traces, or prompt metadata, this one returns savings decisions: what's drifting, what to switch, what to approve, what to audit.
See it in action
Details
- Repository
- github.com/tessera-llm/mcp-server
- Website
- tesseraai.io/dev
- Category
- AI & LLM Tools
- Package
@tessera-llm/mcp-server- Transport
- STDIO
- Auth
- Required
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Mcp Serveragentled/mcp-serverIntelligent AI workflow orchestration with long-term memory, 100+ integrations, and unified credits.0stars
Adaptive Recallaiappsapi/adaptive-recallAdaptive memory system with cognitive scoring, knowledge graph, and self-improving ML.3stars
Risk Deskiman/agent-driven-risk-desk-and-skillsPortfolio risk from supplied data or a synthetic example: VaR, ES, stress, credit, FRTB.0stars
Memxusgpitrella/memxus-remote-mcpPersistent memory layer that saves and recalls your project context and preferences.0stars
UnifAPIunifapi-agent/unifapi-mcp-serverHosted MCP server for live public-data APIs and Skills for AI agents.3stars