keel

# Keel [](https://github.com/angelkurten/keel/actions/workflows/ci.yml) [](https://pypi.org/project/keel-context/) [](https://angelkurten.github.io/keel/) **Stop rebuilding the pipeline for every agent.** Docs: [angelkurten.github.io/keel](https://angelkurten.github.io/keel/) · Why: [We built four agents and four copies of our customer context](https://angelkurten.com/blog/keel-four-agents-four-copies-of-context) Keel is a context layer for AI agents inside vertical SaaS products. Domain events from your product go in; a bitemporal context graph per end-customer comes out; every agent you build reads and writes it through the same five MCP tools. You describe your vertical once, in a *cartridge* (ontology, event catalog, semantic rules, memory policies). Keel's core knows nothing about your domain. Your second agent connects in a day instead of weeks, because there is no second pipeline to write. > Status: 0.11.1, alpha. The MCP server (`keel serve`), the Postgres table adapter and the webhook adapter run against the in-memory store, SQLite (`--db keel.db`, stdlib, no server to run) or MySQL (`--db mysql://...`).  ## The problem Every new agent in a SaaS product repeats the same work: extract the customer's data, normalize it, decide what "critical" means, keep its own memory, expose its own tools to the model. Four agents, four copies of the customer's context, four definitions of "critical", four memories that never talk. Existing memory layers (Zep, Mem0, AgentCore Memory, Neo4j's MKG) ingest text and documents and model memory per user or per agent. None of them start from **domain events**, and none model the **tenant of the tenant**: the end-customer of a B2B SaaS. ## How Keel is different 1. **Cartridges.** A vertical is declared in YAML plus a few ten-line handlers. The core stays generic. Write a cartridge for fleets, clinics, logistics, or your own product. 2. **Events in, not documents.** Keel ingests structured facts with time (Postgres tables, webhooks, Kafka later). Ingestion is deterministic and cheap. The graph is exact because it is born from events, not inferred from text. 3. **Every fact carries its provenance.** Source, `valid_from`, `valid_to`, `recorded_at`, and who corrected it. Audit-ready by construction. 4. **Tenant of the tenant.** One subgraph per end-customer, with its own permissions and expiry. ## Five tools, any agent ``` context(subject, as_of?, depth?, as_known_at?, task?) -> what is true now, at as_of, or as known at a past moment; rules and their verdicts; depth=1 adds the neighbors history(subject?, ..., task?, changed_since?) -> what was true and when we learned it; without a subject, what changed since a watermark search(text?, kind?, rule?, limit?, after?) -> text search, or every entity of a kind / where a rule fires, paged annotate(subject, predicate, value, scope, evidence, task?, key?) -> an agent (or a person) writes what it learned; key makes retries safe related(subject, relation?, as_of?) -> neighbors by relation type ``` Expose them over MCP (`keel serve`) and point any agent framework at them. Declare `agents:` in the cartridge and `keel serve --agent <id>` binds a server to one of them: what it may read and write is enforced by the layer, not by the agent. ## Quick start ```bash uv sync --extra dev uv run keel validate cartridges/fleet --events examples/events/fleet.jsonl uv run keel ingest cartridges/fleet examples/events/fleet.jsonl uv run keel context cartridges/fleet examples/events/fleet.jsonl unit:MXK-4821 ``` Then write your own cartridge: see [docs/first-cartridge.md](docs/first-cartridge.md) (20 minutes). In your own product it is a library: `uv add keel-context` (or `pip install keel-context`); the import is `import keel`. From another language, generate a typed client from your cartridge: `keel sdk typescript|python|go|php cartridges/fleet -o <file>`, with a small runtime per language (see [docs/sdks.md](docs/sdks.md)). ## Connect a real event stream Point an MCP client at the five tools: ```bash uv sync --extra mcp uv run keel serve cartridges/fleet --tenant acme-logistics --events examples/events/fleet.jsonl ``` With Claude Code, one command registers it and the agent sees exactly five tools: ```bash claude mcp add keel -- uv run --directory $PWD keel serve cartridges/fleet --tenant acme-logistics --events examples/events/fleet.jsonl ``` Codex CLI (OpenAI) works the same way from `~/.codex/config.toml`; the four read tools are annotated read-only, so it runs them without prompting. Both clients answered the same questions through the same five tools: see [connect](docs/connect.md#serve-the-tools-over-mcp). A backend, or many tenants from one process, use the same five as a JSON API: `keel serve --transport http` (`/v1/context`, `/v1/events`, ... with OpenAPI; tenant and agent per request). Poll an outbox table in your Postgres (a cursor file makes restarts resume, not re-ingest): ```python import psycopg from keel.adapters.postgres import PostgresTableAdapter adapter = PostgresTableAdapter(lambda: psycopg.connect(DSN), table="keel_events", cursor_path=".keel/cursor") ingestor = Ingestor(cartridge, SqliteStore("keel.db")) # durable; MemoryStore() for a throwaway graph ingestor.ingest(adapter.follow(interval=5)) ``` Or consume a bus: `KafkaAdapter.connect(...)` and `SqsAdapter.connect(...)` in `keel.adapters.bus` (`--extra kafka`, `--extra sqs`), at-least-once, acknowledged after the handler ran. Or receive envelopes over HTTP (`uv sync --extra webhook`): ```python from keel.adapters.webhook import build_app app = build_app(ingestor, token="shared-secret") # POST /events, Authorization: Bearer <token> ``` The table shape and the envelope fields are in `keel/adapters/postgres.py` and `keel/envelope.py`. `keel export <tenant> --db keel.db` writes one tenant's whole audit trail as JSONL, closed facts and provenance included. ## Architecture ``` your SaaS (Postgres) webhooks Kafka / SQS (later) \ | / v v v +-------------------------------------------------+ +------------------+ | event envelope: tenant, type, subject, time | | cartridge | | ingestor + handler registry | <--- | ontology | | bitemporal context graph, one subgraph/tenant | | event catalog | | memory scopes: global / tenant / task / agent | | handlers | | five MCP tools | | semantic rules | +-------------------------------------------------+ | memory policies | | | | +------------------+ agent A agent B agent N (no pipeline) ``` Design rules the core enforces: - **Events are the truth.** An append-only log; the graph is derived and can be replayed. - **Pointers, not copies.** The graph holds references to your rows, the semantics, and earned memory. Never a copy of your tables. - **Explicit time.** Facts close, they are not overwritten. - **Scopes, not a flat store.** Nothing is promoted to a wider scope by accident. ## Repository layout ``` keel/ core: envelope, handlers, store, scopes, cartridge loader, ingest, tools, cli cartridges/ fleet (reference), crm, finance and clinics: four domains, one core, no core change between them examples/ sample event streams and a minimal agent docs/ site sources (mkdocs.yml at the root): first-cartridge, connect, SPEC tests/ ``` ## Roadmap - [x] Event envelope and handler registry - [x] In-memory bitemporal store - [x] Memory scopes with expiry - [x] Five tools as plain functions - [x] MCP server (`keel serve`) - [x] Durable store: SQLite (Graphiti still planned for graph-native queries) - [x] Postgres (table + polling) and webhook adapters - [ ] Consolidation and decay jobs - [ ] Cartridge catalog ## License Apache 2.0. The event envelope, the cartridge format and the five tools are the standard; they will always be free.