What makes this different from a memory tool?

Every other tool retrieves context and trusts the model to use it correctly. Estelle retrieves too, then checks the answer against your real code before it ships. If it references an API that isn't there, it's blocked. Retrieval makes a model better informed; verification makes it reliable.

How can hallucinations be zero?

It is zero under a specific, checkable definition, and it is not a claim that a model never hallucinates. What counts as one: a code reference naming a repo symbol or import your codebase does not define. What the gate does: parses the change with tree-sitter and resolves every such symbol against your repository's real symbol graph, deterministically, with no model call and no network. The rate is candidate-level, per code reference emitted, not per task. Measured by eval_hallucination.py over our own repo and eval_hallucination_libs.py over 1,323 real third-party symbols; the 20-edit figure is a separate run on a private repository no model had seen. In that Python-scoped eval, 0 invented APIs survived. Excluded: references in the 11 navigate-only languages are not gated at all and are left out of the rate rather than counted as passes. What the gate cannot verify: that the code is correct, or that an API behaves as you expect. Python is the only language with the full guarantee of existence, arity, type and member calls on any receiver; 12 of 23 languages block, and the rest enrich the graph without gating an answer.

Is this a memory tool or a coding agent?

Neither. It's the infrastructure both run on. Estelle gives any coding agent whole-repo memory, then gates its output against your real code, runs your tests, and blocks a bad merge. It competes with the model's guessing, not with your editor.

Which tools does it work with?

Anything that speaks the OpenAI API or MCP: OpenCode, Cline, Cursor, Aider, Continue, curl, or your own code. Point it at api.fatelabs.ca/v1 and add your key. It also runs as a language server in any LSP editor (Neovim, Helix, Sublime, Zed, VS Code) and as the %estelle magic in any IPython notebook (Jupyter, Colab, Kaggle).

Can I use my existing Claude, ChatGPT, or Kimi plan?

Yes. Estelle is MCP-native: add it inside the tool where your plan already lives, Claude Code, Cursor, or any MCP client, and your subscription keeps doing the reasoning while Estelle supplies grounded memory and the verification gate underneath. Prefer raw API? Bring a key from Claude, GPT, Kimi, GLM, DeepSeek, or any OpenAI-compatible endpoint instead. Either way, Estelle never marks up a token.

Does it work on private repositories?

That is where it helps most. On code no model has seen before, a stock model starts from zero. Estelle grounds every answer in your actual repo and refuses to guess past it.

Do you train on our code, or store it?

Estelle stores ingested code and session context to provide memory. It does not train a model on your code. When you use a connected model provider, that provider processes the request under its terms. Review the privacy policy and your workspace’s retained sources; memory storage and model training are different operations.

How does an agent hold three million tokens of context?

Estelle retains session context and retrieves a smaller, cited briefing for the next step. The amount recalled depends on the stored evidence, retrieval coverage, and the model’s context limit. A short briefing does not prove every earlier decision was preserved.

How does pricing work?

You bring your own provider key or plan and pay them for tokens at their normal rate. Estelle bills only for the memory-and-verification layer, either a monthly plan or pay-as-you-go credits. We never mark up a token.

checkout.py

1 def checkout(payload):
2~    address = payload.get("address", "").strip()
3+    if not address:
4+        return {"status": 400, "error": "Address required"}
5     return {"status": 200, "address": address}
6 
7 
8 import unittest
9 
10 
11 class CheckoutTests(unittest.TestCase):
12     def test_missing_address(self):
13         self.assertEqual(checkout({}), {
14             "status": 400,
15             "error": "Address required",
16         })
17 
18     def test_blank_address(self):
19         self.assertEqual(checkout({"address": "   "}), {
20             "status": 400,
21             "error": "Address required",
22         })
23 
24     def test_valid_address(self):
25         self.assertEqual(checkout({"address": "12 Market St"}), {
26             "status": 200,
27             "address": "12 Market St",
28         })
29 
30 
31 if __name__ == "__main__":
32     unittest.main()

$ node --test example.test.mjs

ok · default button ok · primary button ok · disabled button ok · both states survive ok · false states add no classes

# pass 5 · # fail 0

$ node --test example.test.mjs

ok · parses a readable duration ok · backs off from numeric milliseconds ok · caps at one minute ok · rejects invalid durations ok · rejects invalid attempts

# pass 5 · # fail 0

POST /checkout · KeyError: 'address' checkout.py:2 in checkout address = payload["address"].strip()

Local Python checks

✓ missing address → 400 ✓ blank address → 400 ✓ valid address → 200

estelle · context

Pick up the button helper. Keep both variants when primary and disabled are true.

estelle_resume · public-clsx-example

WHERE: via claude-code TASK: Build a button helper DECIDED: - We decided: Use clsx object conditions; keep primary and disabled independent, including when both are true. OPEN: - Next: cover the behavior with regression tests.

Parse a retry delay. Keep exponential backoff under one minute.

find_definition · parseDuration · vercel/ms

No definition for parseDuration in the captured source.

index.js:48 defines parse internally. Call the exported ms(value); retry policy belongs in our helper.

Monitor · 2 local exceptions grouped

The request has no address field. Return a 400 response for missing or blank addresses; keep valid checkout working.

Proposed change only. No deployment or pull request was created.

estelle indexacme/checkout

repoacme/checkout4,812 files · code graph

slack#eng-checkout1,340 messages · cited memory

linearENG-214 → ENG-90288 issues · cited memory

pull requests412 merged2,908 reviews · cited memory

productionacme-checkout-prod30d traces · cited memory

0source files in the code graph, mapped to definitions and reference sites

find_definitionissue_tokenauth/tokens.py:8

blast_radiusapp/store.py14 files downstream

verifyrequest.retryWithJitter()not in your repo

agent.tstools.tsone POST to wire it in

01// your agent, unchanged. Estelle does not wrap it

02const run = await agent.run(task);

04// one POST: the stream it just produced, grounded

05await fetch("https://api.fatelabs.ca/agent/events", {

06 method: "POST",

07 headers: {

08 authorization: `Bearer ${process.env.ESTELLE_API_KEY}`,

09 },

10 body: JSON.stringify({ events: run.events }),

11});

13// tool calls, errors and outputs go to memory; a failure

14// binds to its source symbol and opens a reviewable PR

terminalacme/checkout · dev

The agent you already wrote, with the gate inside it. The invented call never left your machine.

acme-storefront.com/checkoutoperational

Acme Storefront$48.00

Emailalex@rivera.dev

Card5312 6698 4471 2035

Pay $48.00

checkout error rate0.1%

estelledeployed with acme/checkout

caught429 rate_limited · POST /v2/charges

grouped1,204 of 3,540 sessions in 20 min, all on the retry loop

read Stripe's live docs429 now requires Retry-After

traced toapi/charge.ts:52

sandbox · a clone, never production6 tests green on the fix

acme/checkout #482open · awaiting review

fix(api): honour Retry-After and reset backoff between attempts

estelle/merge-gategrounded · 6 tests green

checkout is healthy

code graph from the repo · cited memory beside it

Your agents. Your code. Estelle.

Build deterministic agents with Estelle so they don't fail in production.

Stop explaining the same codebase every session. Give your agents the code, decisions, and production context they need. See the evidence behind a proposed fix, then decide what happens next. Keep your editor and your model.

Start free Read the docs

install estelle

Free tier, no card. The gate is not plan-gated: paying raises how much of your codebase Estelle holds, never what it is allowed to do.