# past.dev

> Memory API for AI agents. Ingest timestamped text and query its current state.

The Memory API is in early access. Available features and planned work are documented separately: the reference covers what is live, the roadmap what is not.

## When to use past.dev

Reach for past.dev when software has to answer from unstructured history rather than from documents you control:

- An agent that must know the current state of an account, a project or a person from emails, meeting transcripts, tickets or CRM notes, with the evidence behind each claim.
- Long-running work where facts change over time: past.dev returns which fact currently holds, with prior values dated and sourced, instead of whatever text resembles the question.
- Anything where a wrong answer is worse than no answer: /answer abstains explicitly and every response reports its own cost in dollars.

past.dev is not a web search or a static-docs retriever: it answers only from the history you ingested. There is no schema to define, no embedding model to choose and no chunking pipeline to build.

## How to call it

- REST under https://api.past.dev/api/v1, bearer auth (keys start with past_sk_). Machine-readable surface description: [OpenAPI](https://past.dev/openapi.json).
- MCP (Streamable HTTP) at https://api.past.dev/mcp, manifest at [/.well-known/mcp.json](https://past.dev/.well-known/mcp.json), OAuth sign-in, no key to paste.
- Keys are issued by hand during early access: request access at https://cal.com/team/revo/30min and say what you are building.
- Docs over MCP, no auth: https://past.dev/mcp (server card at [/.well-known/mcp/server-card.json](https://past.dev/.well-known/mcp/server-card.json)). NLWeb queries: POST https://past.dev/ask.
- How agents authenticate: [/auth.md](https://past.dev/auth.md). Machine-readable pricing: [/pricing.md](https://past.dev/pricing.md).
- API discovery per RFC 9727: [/.well-known/api-catalog](https://past.dev/.well-known/api-catalog); agentic resources at [/.well-known/ai-catalog.json](https://past.dev/.well-known/ai-catalog.json).
- Every page answers `Accept: text/markdown`, a `.md` suffix (/index.md for the root), and `?mode=agent`.

## Full documentation in one file
- [All docs as Markdown](https://past.dev/docs/llms-full.txt)

## Pages
- [Memory API for AI agents](https://past.dev): Ingest timestamped text and query its current state, with source evidence.
- [Documentation](https://past.dev/docs): Developer documentation for the Memory API and MCP server.
- [Pricing](https://past.dev/pricing): Free tier, metered on usage, enterprise for dedicated projects.
- [Benchmarks](https://past.dev/benchmarks): How past.dev measures memory: BEAM, LoCoMo on corrected keys, LongMemEval, with baselines, judges, cost and latency.
- [About past.dev](https://past.dev/about): What past.dev is, what makes it different, how the claims are measured, and where the Memory API stands today.
- [Contact past.dev](https://past.dev/contact): How to reach the team: book a call for early access, what to tell us, and where the machine-readable surfaces live.
- [Security](https://past.dev/security): SOC 2 Type II, ISO 27001, ISO 27701, GDPR: the certifications, the controls behind them, and how to get the reports.
- [past.dev brand guidelines](https://past.dev/brand): The marks, colors, type, and usage rules, with downloadable SVGs.
- [Privacy](https://past.dev/privacy): What the site and the API store, who can see it, and how to have it deleted.
- [Changelog](https://past.dev/changelog): What shipped on the Memory API, the MCP server and this site, and when.
- [Context engineering for AI agents](https://past.dev/context-engineering): Context engineering is deciding what goes into the model's context window on every call. How it differs from prompt engineering, what competes for the window, the four moves that manage it, and where memory fits.
- [Vector database for RAG: how to choose one](https://past.dev/vector-database-for-rag): A buying guide for RAG retrieval: pgvector, Pinecone, Qdrant and Weaviate, what each is genuinely best at, the five questions that decide it, and the failure modes no vector database fixes.
- [Vector database vs graph database vs memory](https://past.dev/vector-database-vs-memory): Vector, graph and relational databases answer what resembles a question, how records connect, and which rows match. None answers which fact is true now. A four-way comparison, with the case where memory is the right tool.
- [What is agent memory?](https://past.dev/what-is-agent-memory): Agent memory is what an AI agent knows across sessions, and the mechanism that puts the right part of it back in the context window. The problems it solves, the vocabulary, the existing approaches and where each one stops.
- [Knowledge graph for LLM applications](https://past.dev/knowledge-graph-for-llm): What a knowledge graph gives an LLM, what it costs to build, and the dimension a plain graph leaves out. Multi-hop questions, provenance, entity resolution, and why time changes the model.
- [past.dev vs mem0](https://past.dev/vs/mem0): How past.dev and mem0 differ: extracted memories in a store versus a temporal graph, what each takes in, how each handles a fact that changed, and which one fits your system.
- [past.dev vs Zep](https://past.dev/vs/zep): Zep and past.dev both put time on the fact. Where they differ: the open-source engine, the database underneath, what ingestion takes in, and what the read surface returns.
- [past.dev vs Letta](https://past.dev/vs/letta): Letta is an agent runtime whose agents edit their own memory. past.dev is a memory API with no runtime. Where the two differ, and why they are more often combined than compared.
- [past.dev vs Supermemory](https://past.dev/vs/supermemory): Supermemory is broad context infrastructure with connectors and multimodal ingestion. past.dev is narrow: timestamped text, current-state answers, evidence and abstention. Where each fits.
- [Overview: What the Memory API is](https://past.dev/docs/memory-api/overview) ([Markdown](https://past.dev/docs/memory-api/overview/llms.txt)): Send past.dev your data and ask questions about it.
- [Quickstart](https://past.dev/docs/memory-api/quickstart) ([Markdown](https://past.dev/docs/memory-api/quickstart/llms.txt)): Five calls, about five minutes: push a fact, push its update, wait for the feed to settle, ask two questions.
- [Authentication](https://past.dev/docs/memory-api/authentication) ([Markdown](https://past.dev/docs/memory-api/authentication/llms.txt)): One bearer key per organization. The project is the boundary of a memory.
- [How it works](https://past.dev/docs/memory-api/how-it-works) ([Markdown](https://past.dev/docs/memory-api/how-it-works/llms.txt)): Four properties that affect how to call it.
- [API reference: Endpoint reference](https://past.dev/docs/memory-api/api-reference) ([Markdown](https://past.dev/docs/memory-api/api-reference/llms.txt)): Five endpoints under /api/v1, and what is coming. Every field below is the wire name.
- [Errors](https://past.dev/docs/memory-api/errors) ([Markdown](https://past.dev/docs/memory-api/errors/llms.txt)): A JSON body with code, status, debugMessage and data. Match on the code.
- [Self-hosting](https://past.dev/docs/memory-api/self-hosting) ([Markdown](https://past.dev/docs/memory-api/self-hosting/llms.txt)): The whole thing on your infrastructure with Docker Compose. Same engine, same images as our cloud.
- [Roadmap](https://past.dev/docs/memory-api/roadmap) ([Markdown](https://past.dev/docs/memory-api/roadmap/llms.txt)): What is not built yet, and when it is expected.
- [Overview: What the past.dev MCP server is](https://past.dev/docs/mcp/overview) ([Markdown](https://past.dev/docs/mcp/overview/llms.txt)): A single secure address that puts your own past.dev account in front of an MCP assistant, so you can work on your integration from the assistant you already work in.
- [Quickstart: Installing it, manually](https://past.dev/docs/mcp/quickstart) ([Markdown](https://past.dev/docs/mcp/quickstart/llms.txt)): One URL, no API keys. You'll sign in through a normal browser window, exactly like logging into past.dev.
- [Tool reference: What the assistant can do](https://past.dev/docs/mcp/tool-reference) ([Markdown](https://past.dev/docs/mcp/tool-reference/llms.txt)): Two tools over your own past.dev memories. Each one is scoped, and each one runs as you.
- [Serving your own users: Giving your own users memory](https://past.dev/docs/mcp/serving-your-own-users) ([Markdown](https://past.dev/docs/mcp/serving-your-own-users/llms.txt)): The server above is for your account. To put past.dev memory inside your product's assistant, you run an MCP server of your own over the Memory API. This is the supported shape, and it is deliberately the one that leaves your user directory with you.
- [How it works](https://past.dev/docs/mcp/how-it-works) ([Markdown](https://past.dev/docs/mcp/how-it-works/llms.txt)): past.dev validates who you are, works out what you're allowed to do from its own records, and shows your assistant only the tools that survive that check.
- [Security](https://past.dev/docs/mcp/security) ([Markdown](https://past.dev/docs/mcp/security/llms.txt)): The short version: no long-lived secrets exist to be leaked, authorization is never taken on trust from a token, and a workspace is never a parameter your assistant can change.
- [Admin controls: Administrative controls](https://past.dev/docs/mcp/admin-controls) ([Markdown](https://past.dev/docs/mcp/admin-controls/llms.txt)): What a workspace administrator can govern today, stated plainly, including what is not yet configurable, and the control that compensates for it.
- [Privacy & compliance: Privacy and compliance](https://past.dev/docs/mcp/privacy-compliance) ([Markdown](https://past.dev/docs/mcp/privacy-compliance/llms.txt)): The MCP server is part of the past.dev platform and sits inside the same certified information security and privacy management systems. It is not a side channel around them.
- [Limits: Limits and known behaviour](https://past.dev/docs/mcp/limits) ([Markdown](https://past.dev/docs/mcp/limits/llms.txt)): Things that are true by design, so they don't surprise you later.
- [Troubleshooting](https://past.dev/docs/mcp/troubleshooting) ([Markdown](https://past.dev/docs/mcp/troubleshooting/llms.txt)): Almost every failure is one of four things.
- [FAQ: Questions](https://past.dev/docs/mcp/faq) ([Markdown](https://past.dev/docs/mcp/faq/llms.txt)): The things people ask before they connect it.

## Blog

- [Making a website AI agents can actually read](https://past.dev/blog/website-agents-can-read) ([Markdown](https://past.dev/blog/website-agents-can-read.md)) (2026-08-26, Engineering): The llms.txt files, Markdown mirrors, MCP server and tests behind a 100/100 agent readiness score, and what each one is for.
- [Measuring memory honestly](https://past.dev/blog/measuring-memory-honestly) ([Markdown](https://past.dev/blog/measuring-memory-honestly.md)) (2026-08-25, Research): How past.dev benchmarks memory: corrected answer keys, strong baselines, multiple judges, and cost reported next to accuracy.

The present is the advancing edge of the past.