Privacy Sign In Contact Us
ProtoMap · by HoundDog.ai

ProtoMap: a deterministic service graph.

ProtoMap is a deterministic service graph that maps every service, API and field name across your entire organization's code. It gets updated with every commit and does not need any AI tokens. It plugs into your org’s CI using standard CPU runners.

Install ProtoMap Free Book a Demo
▶ Watch the MozFest demoSee the benchmark →
Updated with every commit
No AI tokens needed
Same graph for engineers and agents, via MCP
6.4×

faster per prompt with the ProtoMap MCP server.

Claude Code with Opus 5.5 and Sonnet 5.5 · average over 20 promptsSee the benchmark →
5.3×

cheaper per prompt, at the same accuracy.

Claude Code with Opus 5.5 and Sonnet 5.5 · average over 20 promptsSee the benchmark →
100%

accuracy with ProtoMap, against 98% without.

Graded against an answer key that does not come from ProtoMap
8×

reduction in tokens and time for change-planning tasks.

Enterprise customer · multi-million-line monorepo
See it in action

Watch ProtoMap answer the questions agents get stuck on.

HoundDog.ai ProtoMap live demo at MozFest 202610:53
Live demo · MozFest 2026HoundDog.ai ProtoMap Demo at MozFestLive demo of HoundDog.ai’s ProtoMap at MozFest’s Open Source House: the service graph, callers and callees, and the blast radius of changing a field name.
Claude Code alone vs. Claude Code with the ProtoMap MCP server: 7x faster, 36x cheaper on one prompt1:09
Benchmark videoProtoMap × Claude Code, side by sideThe same prompt in two live sessions: ProtoMap’s answers in 50 seconds for $0.40, the baseline’s in 6 minutes for $14.44.
Built for the age of agentic coding

gRPC sprawl is coming. Stop burning tokens to rediscover it.

Every line an agent writes adds services, calls and fields that the next agent has to rediscover. ProtoMap keeps one deterministic map of all of it, so your context stays accurate and affordable however fast the code grows.

Monorepo-ready

Built for multi-million-line monorepos and sprawling microservices: one service graph across every repo, refreshed on every commit.

Enterprise-ready

Cloud hosted or deployed on-premises, with GitHub, Bitbucket and GitLab integrations and scans in CI on every pull request.

Future-proof against token costs

Agents query the graph instead of re-reading code, so the cost of an answer depends on the question, not on the size of your codebase.

Ask ProtoMap

The questions behind every cross-service change, answered from code.

Engineers ask in plain English in the web platform. Coding agents ask over MCP. Both get the same deterministic answer.

Callers
Callees
5 hops each way
Click any screenshot to enlarge
How it works

One service graph. Engineers plan with it, agents build with it.

For humans

Plan the change with the full picture

  1. 1

    Install the self-hosted platform

    The self-hosted installation includes a web platform that runs on localhost:3300 and the CLI scanner. You need Docker with Compose 2.23.1 or newer.

    Terminal
    git clone https://github.com/hounddogai/protomap.git
cd protomap
./self-hosted/scripts/install.sh

    Installation guide →

  2. 2

    Scan every repo you need to track

    Run protomap scan against each repository.

    Terminal
    protomap scan /path/to/repository
  3. 3

    Plug in your own AI provider

    Add your AI provider’s API key in the web platform, such as Anthropic, OpenAI or Google, and query the service graph directly in plain English. The assistant answers from the graph and never parses the actual code.

  4. 4

    Plan the change

    Understand the blast radius of changing an API, RPC, method or field, or track service owners and dead RPCs. Get answers across multiple hops, spanning callers, callees and every call site, with visualizations of the graph.

Free for local use. Completely free for whatever code is scanned on the developer’s machine.
For AI coding agents

Apply the change without the discovery

  1. 1

    Connect your coding agent

    Once an engineer is ready to apply a change, the coding agent pulls the dependencies directly from ProtoMap’s service graph over MCP. For Claude Code, run this from the repository’s folder:

    Terminal
    claude mcp add protomap -- protomap mcp serve

    Commands for Codex, Gemini CLI, Copilot CLI, VS Code, OpenCode and more →

  2. 2

    Your changes, laid over the graph

    While the agent runs, the CLI scans the repository as files are saved, so the agent sees your uncommitted changes laid over the shared graph.

  3. 3

    Spend tokens on the change, not on discovery

    The agent no longer greps and reads files across repos to rediscover dependencies. It can focus on the actual change.

Works with every major AI coding agent

Any agent that implements the MCP protocol.

Claude CodeCursorCodexGitHub CopilotOpenCode+ Many others

Code stays inside your environment

Free tier runs on the developer's machine. Enterprise runs in your tenanted cloud or fully on-premises. Your code never leaves your infrastructure either way.

Built in Rust

The engine is extremely lightweight and fast. Built in Rust, it can analyze millions of lines of code in seconds.

Scales with complexity

The larger and more complex the codebase, the more valuable ProtoMap becomes. It thrives where other tools struggle.

Start on one laptop, roll out org-wide

For “human” engineers first. Then for every AI coding agent.

Step 1 · Free

Install it on your machine

Install the ProtoMap web platform and CLI and query your organization’s service graph in plain English. Free for whatever code is scanned on your machine.

  • Self-hosted platform and CLI scanner
  • Bring your own AI provider: Anthropic, OpenAI, Google and others
  • MCP for your coding agent, with your local changes laid over the graph
$ curl -fsSL https://protomap.ai/install.sh | sh Install ProtoMap Free
Step 2 · Enterprise

Keep the graph fresh org-wide

The enterprise version keeps the service graph fresh across your entire organization, so agents stop cloning repositories they don’t own.

  • Cloud hosted or deployed on-premises in your org
  • Direct integrations with GitHub, Bitbucket, and GitLab
  • Runs scans in CI on every pull request
  • HoundDog.ai’s MCP Server gives every agent the full service map
Book a Demo
Static analysis discovers. AI reasons. MCP delivers.

Why not just use an LLM?

HoundDog.ai is built to make your AI more effective, not to replace it.

01

Spend tokens on reasoning, not discovery.

The scanner discovers every service, API and field dependency, so your AI tokens go to higher-value work like assessing the blast radius, planning changes, and answering questions about your service graph.

02

Evidence you can stand behind.

The same commit always produces the same service graph. Every engineer and every agent gets the same answer about what a change will break, instead of output that varies with the prompt.

03

Easy to roll out org-wide.

The Rust scanner runs in CI on standard CPU infrastructure, across every repo and every commit, with negligible impact on build times.

FAQ

Frequently asked questions

What is ProtoMap?+

ProtoMap doesn’t replace your coding agent. It gives your agent the context it is missing. A deterministic static-analysis engine maps every service, API and field across your monorepos and microservices, refreshes the map on every commit, and serves it to engineers in a web platform and to coding agents over MCP.

  • Multi-hop service dependencies: see the blast radius of a change up to five hops out.
  • Method dependencies: find who calls an RPC, plus every call site with its file location.
  • Field dependencies: see which services and repos send or receive a field like credit_card.
  • Code owners: see every inbound service, plus the maintainers and dependents to notify.
  • Unused RPCs: find dead RPCs, orphaned calls, and proto messages declared differently in two repos.
What problem does ProtoMap solve?+

AI coding agents are writing more of our code every day, and every line adds services, calls and fields that the next agent has to rediscover. In a large gRPC + Protobuf codebase, a simple question like “what breaks if I change this service, RPC or field?” sends an agent off to grep, find and read files across repos, trying to piece together relationships in code that often isn’t even checked out on the developer’s machine. That’s slow and expensive, the answer is often incomplete, and it changes from one run to the next.

What ROI can I expect from ProtoMap?+

Because agents query the graph instead of re-reading code, the cost of an answer depends on the question, not on the size of your codebase. For one enterprise customer with a monorepo of millions of lines and a complex web of gRPC services, ProtoMap reduced the time and tokens required for change planning by 8×. In our own benchmarks, ProtoMap makes Claude Code 6.4× faster and 5.3× cheaper across 9 gRPC repositories with 5 million lines of code.

Who is ProtoMap for?+

Engineering teams running gRPC + Protobuf at scale, and the AI coding agents working alongside them.

Is ProtoMap the same as the Dataflow Context Engine?+

Yes. The product previously launched as the HoundDog.ai Dataflow Context Engine, and previewed before that as the API Context Engine, is now ProtoMap. Same engine, same roadmap, new name.

Does my code leave my machine?+

The free local version is installed on your own device: scans run on your machine and your code never leaves it. Enterprise is rolled out org-wide: scans run inside your tenanted environment (cloud or on-premises), and only the resulting catalog and findings are centrally accessible. Your code never leaves your infrastructure in either version.

Which AI coding agents are supported?+

Any AI coding agent that implements the MCP protocol, including Cursor, Codex, Claude Code, GitHub Copilot, and OpenCode.

Which API protocols are supported?+

gRPC is supported today, with GraphQL and REST coming soon.

Where can I find the ProtoMap documentation?+

Quickstart, CLI, coding agent setup, CI and self-hosting guides are at docs.protomap.ai.

Stop wasting tokens on discovery.

Install the self-hosted ProtoMap platform and CLI scanner for free, and give your engineers and agents org-wide context.

Install ProtoMap Free Book a Demo