the operational digital twin

Your system,
mapped from production.

Every endpoint, every dependency, every event, cited to a repo and a commit. Auto-synced on a clock. The substrate humans navigate and agents act on, the same way.

we'll only email you twice. no marketing.
code + runtime·any repo, any org·self-hosted first · byo-llm · in your network
◢ system.model / live
last verified 0s ago
11 nodes · 10 edges · conf ≥ 0.4
verifying
§01 · the problem

Nobody on your team can read your system.

Ask anyone what calls billing. They reconstruct it from grep. Ask an LLM and it invents an answer. The architecture diagram is wrong, the wiki is stale, the senior who knew is on PTO. The system as one object does not exist anywhere.

the diagramdrawn for an architecture that shipped 18 months ago
the wikisix months stale, last edited by somebody who has left
the senior who knewon PTO, the day you needed them
the agentguesses from grep and produces a confident lie
the catalogfilled out once at launch, never updated, never trusted
the deployyou can run, but you cannot read

Every agent that touches your system is guessing, because every human that touches your system is guessing.

§02 · reality, not docs

A map of the system that is actually running.

Your architecture diagram is for the system you wrote down. blastmap is the system that is in production. Read from real signals on a clock. Anchored to repos and commits. Hover any node for relationships. Click any node for the receipt.

what we mean by system

A system is small units that should come together. Your repos. Your deploys. Your traces. Your bus. Your k8s state. Your dashboards. Your runbooks. Each one describes one fragment of the same system. None of them are connected to each other today. We do the connecting.

from production

Extractors walk the live system. Traces, eBPF, the bus, the cluster, the manifests.

every element cited

Every endpoint, event, consumer is anchored to a repo, a commit, and the extractor that found it.

current on a clock

The model resyncs continuously. The freshness stamp is on every edge. Stale answers do not pretend.

every environment

Prod, staging, dev, region by region. Same model, different reality. Drift is named, not hidden.

system.map / live
blastmap architecture mapFrontends12Gateway8Commerce15Payments13AI / ML11Platform14Ops & Tooling9Infra · 3rd party13webmobileadminpartner-portalstatus-pagedashboarddocsembed-widgetagent-consolestudiopreview-appkioskapi-gatewaygraphqledge-cdnwafrate-limitersession-svctrace-collectorloadbalancercheckoutordersshippingcartpromocataloginventorypricingreturnswarehousefulfillmentsubscriptionrecommendreviewsdropshipbillinginvoicingsettlementsaudit-logledgerpayoutsdisputesrefundstaxcurrencyloyaltyentitlementsplan-enginefraud-mlllm-routerembeddingssearch-rankerrecsyssupport-botrag-storefeature-storegpu-poolvector-dbdata-lakeauthconfigprovisioningsecretsfeature-flagsidentityrbacnotificationsschedulerjob-runnerevent-busaudit-busadmin-apigraphql-meshmetricslogsalertstracesebpf-probeincident-mgrcost-watcherror-trackerdashboardspostgresrediskafkaclickhouseelasticsearchs3stripesendgridsnowflakeauth0datadogpagerdutyhubspot
service
frontend
shared
infra
integration
recorded demo · live system map
§03 · on-prem, full stop

Nothing leaves your perimeter.

The map of your system is the most sensitive object your system has. blastmap runs in your network on your hardware. We do not host a SaaS that ingests your graph. We do not share your code with another customer. We do not see it ourselves.

01your vpc

Runs in your Kubernetes, your VPC, your bare metal. There is no cloud edition. There is no shared inference layer. There is no path by which your topology lands in our database.

02byo-llm

The inference layer is your choice. Anthropic, OpenAI, Vertex, Bedrock, Ollama, a fine-tuned house model. The prompts and embeddings run wherever you tell them to run.

03oss core

The extractors, the noisy-OR merge, the refusal logic, the scoring. All open source, all auditable. The thing you run is the thing you can read.

§04 · what the model is

Live documentation of your system. Three surfaces, one source of truth.

The model is not a static catalog. It is a live, verified map of how your system actually runs, plus a verified record of every fact about it. Read it three ways: through the assistant in-app, through MCP from any agent, or from code via the API.

for humans

Assistant

An in-app chat over the model. Plain English in, cited evidence out. Backed by the same graph and the same confidence as everything else.

hosted in app · cited answers · refuses when undetermined
See it answer →
for agents

MCP

The Model Context Protocol surface. Claude, Cursor, Cline, any in-house agent gets 12 read-only tools over the live graph.

stdio · sse · 12 tools · grounded answers
See Claude use it →
for code

API

HTTP + SDK. Query the model from a script, a CI hook, a webhook handler. Every response carries provenance and confidence.

REST + SDK · webhooks · subscribe to changes
See a script call it →
Three surfaces. Three audiences. One source of truth. Whether the reader is a human, an agent, or a script, they all see the same nodes, the same edges, the same confidence, the same refusal.
§05 · see the assistant

Ask. The model walks itself, cites its work.

The same substrate that grounds agents grounds you. Plain English in, evidence out, every claim traceable to an extractor and a line. This is one question, fully resolved, in real time.

observatory · assistant
billing · 14:01 utc
/
ask about a service, cost, blast radius, a module bump…
grounded in the live system graphservices · modules · cost · blast radius
receipts not vibes

Every tool call is named, every result cites the extractor that produced it, every confidence is a real number from the merge. The model refuses to answer when the evidence is not there.

the same brain agents run on

The conversation is one surface of the model. The other surfaces are agents doing the operational work end to end, with the same discipline.

§06 · see claude use the model

Plug your agent into the live graph over MCP.

The same answers, accessed by an agent instead of a human. Claude calls one of the 12 read-only tools, the model returns evidence with confidence, the agent composes a grounded reply. No invented edges, no false certainty.

◢ claude / mcp.blastmap
tool calls via stdio
grounded answers · cites confidence · no hallucinated edges
t=0.0s
one install

Add blastmap from the Claude Code marketplace, or point Claude at your published graph over stdio or SSE. Twelve tools registered, ready to query, auth handled.

the agent inherits the discipline

Every tool result carries provenance and confidence. The agent must rank and gate on it instead of pretending all knowledge is equal. Stale answers do not pretend they are current.

§07 · see code call the model

Query the model from a script.

The same graph, the same confidence, available to anything that can make an HTTP call. Subscribe to changes. Trigger your own systems on confirmed edges. Wire your CI to fail PRs that touch a high-confidence boundary.

@blastmap/sdk · TypeScript
typescript
import { blastmap } from "@blastmap/sdk";

// Walk the model from a changed node, ranked by confidence.
const radius = await blastmap.graph.blast_radius({
  source: "repo:acme/billing@a8f3c21",
  maxHops: 3,
  minConfidence: 0.2,
});

for (const hit of radius.nodes) {
  console.log(hit.id, hit.confidence, hit.provenance);
  // > "repo:acme/checkout" 0.94 [openapi, observed-rpc]
}
curl · REST
bash
curl https://graph.acme.dev/v1/blast-radius \
  -H "Authorization: Bearer $BLASTMAP_TOKEN" \
  -d '{
        "source": "repo:acme/billing@a8f3c21",
        "max_hops": 3,
        "min_confidence": 0.2
      }'

# returns {"nodes": [...], "freshness": "14s", "verified_at": ...}
subscribe to changes

Webhook on every confirmed edge change. Trigger your own pipelines without polling.

gate your ci

Fail PRs that touch a high-confidence integration without a migration plan. The model knows the consumers.

feed your own tools

Pipe the graph into your dashboard, your runbook engine, your internal portal. Same data, your shape.

§08 · how it works

Foundation. Observe. Act.

The OSS code graph is the wedge. Wiring runtime turns it into a live twin. Agents act on the twin. Each layer earns the next.

01

Foundation

The OSS code graph. Every cross-service edge across every repo and org, including the invisible ones formal-schema tools miss.

+ 142 nodes / 308 edges · openapi · grpc · imports · call-sites-llm · wraps
02

Observe

Runtime signals fold in. k8s state, traces, eBPF, logs, cost, the bus. Observed reality sits next to declared intent. Edges confirmed or downgraded against ground truth.

~ 274 edges confirmed · k8s · otel · bus · cost · last verified 14s
03

Act

Agents become buildable on the model. Each does one operational job end to end, cites evidence, refuses to guess when the evidence is not there.

↳ blast-review · live ↳ incident · live ↳ cost, drift, integration
◢ blast.radius / live
payments / a8f3c21 → main
◤ HIGH·6 nodes potentially affected·comment posted
radius = 3 / nodes = 6 / conf ≥ 0.2
watching for edits

Blast radius is one query over the model. The query that started everything.

§09 · live signals

The model is awake.

Health, changes, deploys, traffic, cost. The same substrate, refracted into the lens you happen to need. Every chart you usually rebuild from scratch.

◢ fleet.health / live
8 / 142 services shown
api-gateway
18 msok
auth
12 msok
checkout
24 msok
orders
19 msok
payments
22 msok
billing
14 msok
fraud-ml
41 msok
event-bus
9 msok
confirmed via runtime · 4s ago
8 services · 8 ok
◢ changes / live
last 24h · 412 events
deploy
checkout·rolled v412 to prod-eu / 6 pods
@auto
every action the system took, ranked by recency
t=0.0s
§10 · democratize the knowledge

Stop being the only person who can answer.

Anyone on the team can answer what depends on this, what calls billing, what would break, without paging the senior architect. The architect stops being the bottleneck and starts being the architect.

01

The new joiner

Ramps up on the system on day one, without booking time with three different seniors.

02

SRE · on-call

Sees what changed in the window and the most likely cause with cited evidence. Stops the archaeology.

03

Engineering manager

Reads who owns what and where the bus factor is. Without asking the team to fill out another spreadsheet.

04

Platform · DX

Finds every consumer of the API they want to bump. Bounds the blast. Plans the deprecation, does not guess at it.

05

Senior architect

Stops being the bottleneck. Stops being the answer to every 'what depends on this'. The team finds out for themselves.

06

FinOps · cost

Reads cost over the graph, not over a billing CSV. Knows which module rents the GPU pool nobody calls.

§11 · differentiation

Not an AI-SRE. Not a catalog.

The shared thread: provenance and confidence are the discipline that makes the labor safe to sell. An agent that is confidently wrong is worse than none.

versus the ai-sre wave

Resolve · Traversal · Cleric · Causely

They start from zero on every incident.

Runtime-only, reconstruct context per incident, so they guess. blastmap agents run on a durable code-plus-runtime model, so they compound and they cite.

versus catalogs and portals

Cortex · Port · Backstage

A catalog tells you what exists.

You maintain it by hand. blastmap tells you what happens if you touch it, then does something about it. Automatic, verified, and it acts.

Both categories default to SaaS. They ingest your topology into their cloud and read it from there. blastmap runs in your network. We never see your graph.

§12 · five agents

Built on the model. Doing the work, end to end.

Five agents that fall directly out of the model. Each does one operational job, cites its evidence, and refuses to guess when the evidence is not there.

01every pr

Blast reviewer

Walks the live model from the changed node and writes the confidence-ranked comment a senior reviewer would have written. The PR knows what it might break before merge.

pr · open#1247 · payments / tighten cookie scope
blast reviewer scenepaymentsa8f3c21checkout0.94orders0.91billing0.87audit0.72loyalty0.48
02every page

Incident first responder

Page lands and the same brain runs the recon: window, traces, recent deploys, the most likely cause with cited evidence. The on-call reads a verdict, not a smoking pile of logs.

page · 14:09checkout p95 ↑ 8x over baseline
13:3014:00now · 14:09
green-to-red window14 events scanned
03every contract bump

Integration sentinel

Catches the breaking schema change the moment it lands, walks every consumer from the model, blocks the merge if anyone is on incompatible shape, proposes the shim.

contract diffPOST /v2/charge · breaking field removed
schema_v2 {
- idempotency_key: string
+ idempotency_token: uuid
}
walking consumers · 7 found
· checkout· orders· billing· audit· loyalty· settlements· fraud-ml
04every week

Cost cartographer

Sweeps runtime usage against the graph. Names the modules paying for capacity nobody calls, and proposes a new shape the model can verify is safe to apply.

cost sweep · weekly142 services · runtime usage vs declared shape
provisioning$2,261/mo
api-gateway$890/mo
searchok
reporting$1,140/mo
05every minute

Drift watcher

Watches declared shape against observed reality. Flags the moment k8s drifts from the manifest, names the divergence, routes to the team the model says owns the node.

drift checkmanifests vs k8s ground truth
signaldeclaredobserved
provisioning · replicas35
billing · cpu req500m1200m
checkout · ingressv1v1
notifications · imagev412v410
pre-launch · v0.1 incoming

Ship without fear
of the unknown.

Get the v0.1 launch notification and a 30-day head start before public release.

we'll only email you twice. no marketing.