Tempera
Gives Claude Desktop persistent memory across coding sessions through semantic episode capture and retrieval. Instead of forgetting everything between chats, Claude can store what worked, search past solutions, and learn from feedback through a reinforcement learning loop that boosts helpful memories and fades stale ones. Exposes tempera_capture, tempera_retrieve, and tempera_brief tools. The brief surface loads pending corrections, calibration warnings, and relevant patterns before you start a task. Dream cycle runs nightly to detect contradictions and extract reusable templates. Supports cross-project learning with scoped decay rates so universal truths persist while project workarounds expire naturally. Built in Rust with local BGE embeddings, no cloud dependencies.
- GitHub stars
- 4
- tools
- 12
- transport
- STDIO
- last updated
- May 30, 2026
Tools
12 toolsPublic tool metadata for what this MCP can expose to an agent.
tempera_session_startCall ONCE at the very start. Returns any clarifying question tempera drafted after a previous failed/partial session in this project.
Call ONCE at the very start. Returns any clarifying question tempera drafted after a previous failed/partial session in this project.
No parameter schema in public metadata yet.
tempera_briefCall once the file set is known. Joins pending ask-back, reasoning template, top correction categories for these files, should-have-asked triggers, and calibration warning into one response.
Call once the file set is known. Joins pending ask-back, reasoning template, top correction categories for these files, should-have-asked triggers, and calibration warning into one response.
No parameter schema in public metadata yet.
tempera_retrieveSearch for similar past episodes. Set scope="cross-project" to include transferable claims from other projects.
Search for similar past episodes. Set scope="cross-project" to include transferable claims from other projects.
No parameter schema in public metadata yet.
tempera_templatePull the reasoning template stored for a (task_type, domain) pair. The step sequence past wins followed.
Pull the reasoning template stored for a (task_type, domain) pair. The step sequence past wins followed.
No parameter schema in public metadata yet.
tempera_log_correctionWhen the user corrects an assumption / decision / piece of code. Categorized log; the brief surface uses it.
When the user corrects an assumption / decision / piece of code. Categorized log; the brief surface uses it.
No parameter schema in public metadata yet.
tempera_log_should_have_askedWhen you realize mid-task you should have asked a question up front. Records the trigger context, the question, and the eventual answer.
When you realize mid-task you should have asked a question up front. Records the trigger context, the question, and the eventual answer.
No parameter schema in public metadata yet.
tempera_captureSave session as an episode. Auto-detects session links and runs propagation. The intent-extraction LLM call also suggests a ValidityScope for cross-project routing.
Save session as an episode. Auto-detects session links and runs propagation. The intent-extraction LLM call also suggests a ValidityScope for cross-project routing.
No parameter schema in public metadata yet.
tempera_feedbackMark retrieved episodes as helpful or not. Drives the utility-learning loop.
Mark retrieved episodes as helpful or not. Drives the utility-learning loop.
No parameter schema in public metadata yet.
tempera_statusPer-project memory health snapshot.
Per-project memory health snapshot.
No parameter schema in public metadata yet.
tempera_statsStatistics + trend analytics (helpfulness over time, domain growth, learning curve).
Statistics + trend analytics (helpfulness over time, domain growth, learning curve).
No parameter schema in public metadata yet.
tempera_propagateMulti-hop Bellman propagation with convergence tracking. Periodic maintenance.
Multi-hop Bellman propagation with convergence tracking. Periodic maintenance.
No parameter schema in public metadata yet.
tempera_reviewConsolidate similar BKMs, cleanup. Run after related task series.
Consolidate similar BKMs, cleanup. Run after related task series.
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 languageReadme
Tempera - Persistent Memory for Claude Code
Tempera gives Claude Code a persistent memory that learns from experience. Instead of starting fresh each session, Claude can recall past solutions, learn what works, and get smarter over time.
Why Tempera?
The Problem: Claude Code forgets everything between sessions. You solve the same problems repeatedly, and Claude can't learn from past successes or failures.
The Solution: Tempera captures coding sessions as "episodes", indexes them for semantic search, and uses reinforcement learning to surface the most valuable memories when relevant.
Details
- Repository
- github.com/anvanster/tempera
- Category
- AI & LLM Tools
- Package
@astudioplus/tempera-mcp- Transport
- STDIO
- Registry
- MCP registry, active
Related AI & LLM Tools MCP servers
View all
Smart Context Mcparrayo/smart-context-mcpReduces AI agent token usage by 90% via context compression and task checkpoint persistence.4stars
Meyhem — MCP Server Discovery & Agent Searchc5huracan/meyhemDiscover 6,700+ MCP servers and 15,000+ OpenClaw skills. Agent-native search with outcome ranking.4stars
BananaCrystal: FX rates, swaps and payments in 140+ currenciesbananacrystal/mcp-server-bananacrystalGive your AI agent a wallet: live FX rates, swaps and payments in 140+ local currencies.4stars
Feedthroughfeedthrough/feedthroughDebug bridge inside your web app: an agent reads the DOM, console and network. Any browser.4stars
Archivist AIastrotomic/mcp.myarchivist.aiOfficial remote MCP server for Archivist AI TTRPG campaign memory: characters, sessions, and more.0stars
Korext: AI Code Governancekorext/enforce-actionGovernance copilot for AI-assisted coding. 72 packs, 532 rules, proof bundles.0stars