Enables AI coding assistants to perform multi-model debates to reach consensus and provides provenance tracing for software decisions.
Also known as spacepacket1/e3d-pilot
- Stars
- 7
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 0.2.0
- Last commit
- Nov 15, 2026
- License
- MIT
What is e3d-pilot MCP?
e3d-pilot enables AI coding assistants to perform multi-model debates to reach consensus and provides provenance tracing for software decisions.
Summary generated from the project's README
About e3d-pilot MCP server
Developers seeking to avoid single-model errors can use e3d-pilot to facilitate multi-model debates within their coding workflows. By running several configured models like Claude, Codex, Grok, or Devin through the e3d_debate tool, the system forces independent answers, critiques, and revisions to reach a consensus or highlight specific dissent. This process helps ensure that design decisions are vetted by multiple perspectives rather than relying on a single AI's output.
This MCP server integrates directly into editors like Cursor, Claude Code, and GitHub Copilot using a local stdio setup. It requires Node 18+, jq, and bash 4+, along with existing model provider CLIs or API keys configured in the environment. Beyond debating, the system provides provenance tracing through e3d_provenance_trace, which uses an append-only ledger to reconstruct the history of a software change from its initial idea through execution and review.
What you can do with it
- 1Resolve architectural disagreementsAsk an AI assistant to run a debate between multiple models to decide between two different storage layers.
- 2Identify model dissentUse the synthesizer to find where different AI models disagree on a technical implementation.
Written from the project's README. Every use case is checked against a quote from it.
How to install e3d-pilot MCP
Formats verified 10 Nov 2026Add to CursorAdd to GooseInstall e3d-pilot MCP in Claude Code
$ claude mcp add e3d-pilot -- npx -y e3d-pilotInstall e3d-pilot MCP in Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"e3d-pilot": {
"command": "npx",
"args": [
"-y",
"e3d-pilot"
]
}
}
}Install e3d-pilot MCP in VS Code
.vscode/mcp.json
{
"servers": {
"e3d-pilot": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"e3d-pilot"
]
}
}
}Install e3d-pilot MCP in VS Code CLI
$ code --add-mcp '{"name":"e3d-pilot","type":"stdio","command":"npx","args":["-y","e3d-pilot"]}'Install e3d-pilot MCP in Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"e3d-pilot": {
"command": "npx",
"args": [
"-y",
"e3d-pilot"
]
}
}
}Install e3d-pilot MCP in Codex
$ codex mcp add e3d-pilot -- npx -y e3d-pilotInstall e3d-pilot MCP in Codex config.toml
~/.codex/config.toml
[mcp_servers.e3d-pilot] command = "npx" args = ["-y", "e3d-pilot"]
Install e3d-pilot MCP in Gemini CLI
$ gemini mcp add -s user e3d-pilot npx -- -y e3d-pilotInstall e3d-pilot MCP in Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"e3d-pilot": {
"command": "npx",
"args": [
"-y",
"e3d-pilot"
]
}
}
}Configuration
ANTHROPIC_API_KEYANTHROPIC_API_MODELAWS_REGIONAZURE_OPENAI_API_KEYAZURE_OPENAI_DEPLOYMENTAZURE_OPENAI_ENDPOINTBEDROCK_MODEL_IDDEVIN_BINDEVIN_MODELDEVIN_PERMISSION_MODEE3D_MCP_MAX_CONCURRENTMaximum concurrent debates (default 2).
E3D_MCP_MAX_PROVIDERSMaximum model providers per debate (default 4).
E3D_MCP_MAX_ROUNDSMaximum debate rounds per call (default 3).
E3D_MCP_REPOSE3D_MCP_TIMEOUT_SECPer-debate timeout in seconds (default 900).
E3D_PILOT_TRACKING_PROJECT_URLE3D_PILOT_WEB_AUTH_PASSE3D_PILOT_WEB_AUTH_USERE3D_SPECIALISTSGEMINI_API_KEYGEMINI_API_MODELLOCAL_MODEL_ENDPOINTLOCAL_MODEL_NAMEOPENAI_API_KEYOPENAI_API_MODELSTUB_SERVICE_TOKENXAI_API_KEYXAI_API_MODEL
Automated checks
Ran Nov 13, 2026Full report →Passed our automated checks. That is not a security audit: read the code and give the server the narrowest access it needs. How checks work
- Source repository linked and reachablespacepacket1/e3d-pilot is public and reachablegithub.com/spacepacket1/e3d-pilot ↗Pass
- Package names this repository as its sourcee3d-pilot names spacepacket1/e3d-pilot as sourcenpmjs.com/package/e3d-pilot ↗Pass
- Build provenance (Sigstore attestation)No build attestation published for this releaseNot checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-11-07github.com/spacepacket1/e3d-pilot ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNo popular package one edit away from e3d-pilotPass
- No critical advisory in the package or its dependencies2 resolved dependencies checked; no critical advisoryosv.dev ↗Pass
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHube3d-pilot
Multi-model AI problem solving: independent answers → critique → convergence → human decision → execution → review → provenance.
e3d-pilot is an open-source system for getting multiple AI models to work on the same problem instead of trusting a single model's answer — from a one-off question to a fully executed, human-gated software change.
It starts simply. Ask several configured models (Claude, Codex, Grok, Devin, or others you wire in) the same question. Let them answer independently, see and critique each other's positions, revise, expose real dissent, and converge on the strongest answer. That's bin/e3d-debate — a standalone tool, no repo or pipeline required. It's also available inside your editor as an MCP server: use it from GitHub Copilot, Claude Code, or Cursor with one npx line.
Then e3d-pilot takes the same instinct further, into real software work: discover → ideate → approve → draft → negotiate → execute → review → publish → outcome. Every step is preserved in an append-only ledger, so the system (or a human, or a fresh AI session) can later reconstruct where an idea came from, which models or people influenced it, what it produced, and what happened afterward.
Try e3d-debate in 60 seconds
git clone https://github.com/spacepacket1/e3d-pilot.git
cd e3d-pilot
bin/e3d-debate \
"Should e3d-pilot's provenance data be a graph, or is that over-engineering?" \
--providers claude,codex --rounds 3
(Any lib/providers/<name> adapter can join — add devin/grok-build/others if you have them configured; claude/codex are the two most people trying this will already have.)
Round 1: every provider answers independently. Round 2+: each sees the others' prior-round positions and must defend, revise, rebut, or concede. The debate stops early if they genuinely converge; a synthesizer then reads the full transcript and reports a final answer, consensus, and any remaining dissent.
That question isn't hypothetical — it's a real debate we ran this way, condensed from the full transcript:
QUESTION
Should the E3D 3D graph generalize into a standalone product, or should
e3d-pilot instrument its own provenance instead?
CLAUDE (round 3, the only one to formally reach "approved")
Don't create e3d-graph yet — build a confidence-annotated, domain-neutral
provenance emitter inside e3d-pilot, prove attribution questions are
answerable from the data alone, add a 2D lens only if they are, and extract
a public repo only on external pull or paid demand.
CODEX (round 3)
Build outcome-linked Pilot provenance first, then open-source a narrow
standalone viewer only if measured workflow value and external pull
justify extraction.
DEVIN (round 3)
Instrument Pilot with a domain-neutral provenance schema from day one; gate
public extraction on the internal experiment succeeding, not on indefinite
deferral.
GROK-BUILD (round 3)
Do not ship standalone e3d-graph; instrument Pilot provenance first and
treat visualization as a 2D lens inside E3D.
SYNTHESIS
Do not ship a standalone e3d-graph product. Instrument e3d-pilot now with
an append-only, domain-neutral, confidence-annotated provenance schema —
the ledger stays canonical, the graph is an index, not memory. Validate by
answering real attribution questions from the data alone before any UI.
DISSENT (reported, not hidden)
Devin and partly Codex would gate public extraction on the internal
experiment succeeding; Claude and Grok-build require external pull or paid
demand instead — a real, named disagreement the tool didn't paper over.
What happened next: we built exactly that, then ran it against three real ideas in this repo's own history. All three came back "more useful than reading the raw ledger by hand" — and one of them caught a real error in the spec that proposed it. Later graph-intelligence queries were evaluated just as honestly: one clear win, two explicitly "no signal" because the real data was still too sparse — reported as-is, not oversold. That full arc (spec, dogfood results, the honest misses) is in docs/provenance-graph-experiment.md and docs/graph-intelligence-experiment.md.
Experimental: the provenance graph. e3d-pilot now derives structural relationships — ideas, actors, models, decisions, repos, commits, evidence, outcomes — from its own immutable event ledger. We're testing whether that graph reveals patterns that are genuinely hard to get from the raw ledger, not claiming it already does.
You don't need the rest of e3d-pilot to use e3d-debate on its own — see e3d-debate below for the full flag reference. Keep reading for the part that takes a debate's answer all the way to a shipped, reviewed, provenance-tracked change.
Use e3d-pilot from GitHub Copilot, Claude Code, and Cursor (MCP)
e3d-mcp is a zero-dependency Model Context Protocol server. It lets an AI coding assistant ask for a multi-model second opinion mid-task, instead of committing to the first answer one model gives it. It's published on npm as e3d-pilot and listed in the official MCP registry as io.github.spacepacket1/e3d-pilot.
| Tool | What it does |
|---|---|
e3d_debate |
Several independent models answer and a synthesizer returns a decisive answer, a consensus level (full / majority / split), any named dissent, and each model's final position, as both readable text and structured data (dissent.present, dissent.summary, positions). mode: "quick" (default) is one round and takes seconds; mode: "thorough" adds a critique round where models rebut each other. The full transcript is opt-in (include_transcript). |
e3d_quote |
Hosted per-token servers only. Free and read-only: prices an e3d_debate (same arguments) and returns quote_id, max_credits, expires_at and your key's limits; nothing is held or spent. See Agents. |
e3d_list_providers |
Shows which model providers are available on this machine. A debate needs at least two. |
e3d_provenance_trace |
Read-only: traces why a change exists, from its origin through debates, approvals, execution, review, and outcome. |
e3d_ideas_list |
Read-only: lists the ideas e3d-pilot tracks in a repo, to find an id for the trace. |
e3d_gallery |
Opt-in (E3D_MCP_GALLERY=1; on at pilot.e3d.ai). Free and read-only: searches and reads the public Pilot Gallery, the debates people chose to publish, and returns the question, answer, consensus and any dissent with a link. Nothing is spent or written. |
e3d_list_experts, e3d_get_expert |
Opt-in (E3D_MCP_EXPERTS=1). Free and read-only: describe the built-in experts and presets, and show the exact lens text a seat receives. They only describe: passing skills through the e3d_debate tool is not built. |
A real result in quick mode (11 seconds, two models), from asking whether a small CLI should default to a flat JSON file or SQLite:
FINAL ANSWER: Default to a flat JSON file for a small CLI. It has no dependencies, is
human-readable and diffable, and is easy to hand-edit and debug. Write it atomically
(temp file, then rename). Move to SQLite once you need indexed queries, concurrent
writers, or data beyond a few MB.
CONSENSUS: split
DISSENT: Codex prefers SQLite as the default for any persistent mutable state because
it gives transactions, concurrency safety, and schema evolution at low cost.
The disagreement is the point: you get the answer and the strongest case against it, instead of one model's confident guess.
Install
VS Code (GitHub Copilot agent mode): one-click install, or add this to .vscode/mcp.json:
{
"servers": {
"e3d-pilot": { "type": "stdio", "command": "npx", "args": ["-y", "--no-update-notifier", "-p", "e3d-pilot", "e3d-mcp"] }
}
}
Claude Code:
claude mcp add e3d-pilot -- npx -y --no-update-notifier -p e3d-pilot e3d-mcp
Cursor (~/.cursor/mcp.json) and other clients that use the mcpServers shape:
{ "mcpServers": { "e3d-pilot": { "command": "npx", "args": ["-y", "--no-update-notifier", "-p", "e3d-pilot", "e3d-mcp"] } } }
Any MCP client that can launch a stdio server should work; the shapes above are each client's standard format.
Prerequisites: Node 18+, jq, bash 4+ (macOS ships 3.2: brew install bash jq), and at least two model CLIs logged in (e.g. claude and codex). Check them all at once:
npx -y -p e3d-pilot e3d-mcp --doctor
--doctor reports what's missing and how to fix it. It does not need the full pipeline's e3d-code executor (it only notes whether that is installed). From a clone, bin/e3d-mcp --print-config --local prints a config that points at your checkout.
Using it
In Copilot agent mode (or any client), ask: "Before you pick a storage layer, run e3d_debate on SQLite vs flat JSON for this CLI." To make the assistant do this unprompted for contestable design choices, copy examples/copilot-instructions.md into .github/copilot-instructions.md or AGENTS.md. Don't use it for lookups or questions with an obvious answer; it is slow and costs money.
Cost and limits
quickdebates take seconds to a minute andthoroughones a few minutes; both spend your model credits. The server capse3d_debatewithE3D_MCP_MAX_ROUNDS(3),E3D_MCP_MAX_PROVIDERS(4),E3D_MCP_MAX_CONCURRENT(2), andE3D_MCP_TIMEOUT_SEC(900). When every slot is busy a call waits its turn, first in and first out, for up toE3D_MCP_QUEUE_WAIT_SEC(90);e3d_debate_startqueues without holding the call.- It uses whatever provider CLIs and keys are in the server's environment (see
e3d-pilot providers list). - The Copilot coding agent runs on GitHub-hosted runners, where those provider CLIs aren't installed. Use the local (stdio) setup above for now.
Web search in debates
Every debate prompt states today's UTC date. The API providers (anthropic-api, openai-api, gemini-api, xai-api) can also use their vendor's own web search tool in the debate rounds. It is on by default, and the question text may then be sent to a search provider. CLI voices, MCP-sampling voices and the synthesizer never search; they are told they have no web access. Pass web_search: false to e3d_debate or e3d_debate_start, or --web-search off to e3d-debate, to turn it off for one debate. Search results are treated as untrusted data: see docs/web-search-providers.md for what each vendor allows and what risk remains, and docs/web-search-interface.md for the result shapes.
| Variable | Default | Effect |
|---|---|---|
E3D_DEBATE_WEB_SEARCH |
on |
off is the kill switch: no search anywhere, no code change needed. Any other value also turns search off. |
E3D_DEBATE_WEB_SEARCH_PROVIDERS |
anthropic-api,openai-api,gemini-api,xai-api |
Which providers may search. Names other than these four are ignored. |
E3D_DEBATE_SEARCH_MAX_PER_CALL |
3 (1-10) |
Search cap per model call. It is native for Anthropic and OpenAI, the closest available control for xAI, and checked after the fact for every vendor, with an ALARM when a vendor goes over. |
E3D_DEBATE_SEARCH_MAX_CITATIONS |
8 (1-20) |
Citations kept per call. Extra citations are discarded and counted. |
E3D_DEBATE_SEARCH_EXCERPT_CHARS |
150 (0-300) |
Hard cap on an excerpt kept from a citation. 0 keeps none. |
E3D_DEBATE_SEARCH_ALLOW_DOMAINS |
empty | Comma list of bare domains (example.com). Applied natively where the vendor supports it and post-filtered everywhere. It takes precedence where a vendor accepts only one list. |
E3D_DEBATE_SEARCH_DENY_DOMAINS |
empty | Comma list of bare domains. Native where possible, post-filtered everywhere. |
E3D_DEBATE_SEARCH_TIMEOUT_SEC |
derived | Wall-clock limit for one search call. By default it is the debate timeout divided by (2 x rounds + 1), at least 10 s, or 120 s when no debate timeout is set. |
E3D_MCP_DAILY_SEARCHES |
0 (unlimited) |
Daily cap on web searches per principal. The worst case (cap x searching calls) is reserved before a debate runs. |
GEMINI_API_NATIVE_BASE_URL |
https://generativelanguage.googleapis.com/v1beta |
Endpoint for Gemini's Interactions API, which is used only when Gemini searches. |
OPENAI_API_SEARCH_CONTEXT_SIZE |
low |
search_context_size for OpenAI's search tool. |
ANTHROPIC_API_MAX_PAUSE_CONTINUATIONS |
2 (0-5) |
How many pause_turn continuations a search call may make. Beyond that it fails closed. |
If a search call fails, times out, uses a tool other than search, or the vendor has search disabled, that model is asked again without search, and the answer says so ("web search was unavailable ..."). The answer also reports how many models could search ("on for N of M models").
e3d_debateand the read-only tools never write to a repo. The pipeline tools below do, only through the same gated CLI paths as a person would, ande3d_pilot_proposewrites to the repo's.e3d-pilot/ledger.
e3d-pilot MCP questions
How do I add e3d-pilot MCP to Claude Code?
Run claude mcp add e3d-pilot -- npx -y e3d-pilot in your terminal. Add --scope user to make it available in every project.
How do I add e3d-pilot MCP to Cursor?
Open ~/.cursor/mcp.json and add the server under mcpServers, using the Cursor config shown in the install section above.
What does e3d_debate do?
It allows several independent models to answer a question, critique each other, and converge on a synthesis that includes consensus and named dissent.
How can I use e3d-pilot with my coding assistant?
You can use it with GitHub Copilot, Claude Code, or Cursor by adding the server via npx in your MCP configuration.
Does the debate process include web searching?
Yes, web search is enabled by default for certain API providers like Anthropic, OpenAI, Gemini, and xAI during debate rounds.
What is the purpose of e3d_provenance_trace?
It is a read-only tool that traces the history of a change from its origin through debates, approvals, execution, and review.