What did Hlido score Proliferate?

Proliferate scored 73/100 (STEADY) in Hlido's independent, hands-on review.

How much does Proliferate cost?

Hlido recorded the pricing model as: Open source. Derived from evidence Hlido holds, not vendor-supplied; verify current prices on the vendor's pricing page.

Does any vendor pay Hlido for placement?

No. Hlido takes no money from the agents it rates — scoring weights stay private and the evidence behind every verdict is public.

Proliferate

Name: Proliferate review
Item: Proliferate
Rating: 73
Author: Hlido Editor

Coding · tested 2026-09-01 · by the Hlido desk, not the vendor

In short: Open-source, self-hostable workspace for running coding agents in parallel — a review-first orchestration layer over Claude Code, Codex, Cursor, OpenCode and Grok.

5 PASS · 0 FAIL of 5 public-surface claims

Visit Proliferate → scorecard.json

Quick answer

Proliferate scores 73/100 (STEADY) on Hlido’s independent, hands-on test (reviewed 2026-09-01). STEADY (73) on a medium-confidence public-surface read: the positioning is clear and differentiated (open-source, self-hostable, agent-agnostic, review-first), the docs are well-structured, and the concept maps directly Pricing: Open source (free entry point documented).

Hlido Editor · 2026-09-01

Proliferate is not another coding agent — it is the workspace you run your coding agents inside. The pitch on the public docs is coherent and, for an agentic ecosystem, well-aimed: pick whichever agent fits a task, give it an isolated git worktree with its own branch, terminal, conversation and review surface, and read the diff before anything ships. Chaining agent steps and human approvals into a repeatable workflow is the second act. That 'review before it ships' framing is the right instinct for the current moment, where the bottleneck is trust in agent output, not raw generation. Being open-source (AGPL-3.0) and self-hostable is a genuine differentiator against the hosted-only orchestrators, and agent-agnosticism (it drives Claude Code, Codex, Cursor, OpenCode, Grok) means it doesn't bet the product on one model vendor. What the public surface can't yet settle is execution depth: this is an early project (a few hundred GitHub stars, a Windows build still marked beta), the docs describe the concepts more than they prove the day-to-day reliability of parallel worktrees under load, and there is no visible programmatic/MCP interface for an agent to drive Proliferate itself — it is a human-operated cockpit, not a component another agent composes with. Promising positioning; the reliability and adoption evidence is still accruing.

Why STEADY

STEADY (73) on a medium-confidence public-surface read: the positioning is clear and differentiated (open-source, self-hostable, agent-agnostic, review-first), the docs are well-structured, and the concept maps directly to a real 2026 pain (governing multiple coding agents safely). Held below the VITAL band because this is an early-stage project — a small star count, a beta Windows build, and documentation that explains the model more than it evidences production reliability. Not FADING because there are no trust red flags: the code is open, the license is explicit, and the value proposition is honest about being local and self-hostable rather than over-claiming.

Public-surface checklist

PASS Homepage loads (required)
PASS Primary value prop (required) — 'Run coding agents and build reviewable workflows locally.'
PASS Cta present (required) — 'Download for macOS' / Quickstart
PASS Pricing or access — Open-source (AGPL-3.0), self-hostable — free to run
PASS Evidence or demo — Quickstart + 'How Proliferate works' docs captured

What we saw

4 screenshots captured by the Hlido engine during the reviewed run (run-cb3fe144314e78ea-proliferate-com). Our own captures — not vendor marketing material.

Proliferate — run screenshot 1 (home.png) — home.png

Proliferate — run screenshot 2 (page_quickstart.png) — page_quickstart.png

Proliferate — run screenshot 3 (page_windows.png) — page_windows.png

Proliferate — run screenshot 4 (page_how-proliferate-works.png) — page_how-proliferate-works.png

What it does well

Isolates each agent task in its own git worktree (branch, terminal, chat, review surface) — a clean concurrency model
Agent-agnostic: drives Claude Code, Codex, Cursor, OpenCode and Grok rather than locking to one model vendor
Review-first workflow — read the diff and gate on human approval before anything ships
Open-source (AGPL-3.0) and self-hostable, so the workspace can run entirely on the user's own machine
Documentation is well-organised (Quickstart, Concepts, Features, Reference) and honest about scope

What it fails at

Early-stage maturity signals: small GitHub star count and a Windows build still marked beta
No visible programmatic or MCP interface — Proliferate is a human-operated cockpit, not something another agent can drive or compose with
Public docs describe the concurrency/worktree model more than they evidence its reliability under real parallel load
AGPL-3.0 can be a licensing constraint for teams that need to embed it in closed products

Best for

Developers running several coding agents at once who want each task isolated and reviewable
Teams that require self-hosting and want to avoid a hosted-only orchestration vendor
Anyone standardising a 'read the diff before it ships' review gate around agent output

Not recommended for

Users wanting a single turnkey hosted product with no local setup
Agent-driven pipelines that need to invoke the orchestrator programmatically (no API/MCP surface exposed)
Teams blocked by AGPL-3.0 obligations

Pricing & access

ModelOpen source
Free entry pointYes — a free tier or open-source edition is documented
Pricing findable on the public surfacePASS Open-source (AGPL-3.0), self-hostable — free to run (tested 2026-09-01)

Derived from Hlido-held evidence only (engine checklist + editorial text); quotes are verbatim from the scorecard; not vendor-supplied; re-derived daily. Verify current prices on the vendor's pricing page. Last verified 2026-09-01.

Compared to

OpenCode multi-agent-orchestration
OpenCode is itself a coding agent; Proliferate is the workspace you'd run OpenCode (and others) inside. They're complementary, not substitutes — Proliferate wins when the problem is orchestrating and reviewing multiple agents, not the agent itself.
Aider review-first-workspace
Aider is a mature, single-session terminal coding agent with a large following; Proliferate is a younger orchestration layer above that class of tool. Choose Aider to do the coding; choose Proliferate to marshal several such agents with per-task isolation and review gates.

Agent relevance

CLI

Agentic-Commerce Readiness 43/100 · SURFACE-ONLY

Independent readiness for agent delegation & transaction. How it’s scored · check live

Proliferate orchestrates coding agents rather than exposing itself to them. It is operated by a human (desktop app + local install) and does not publish an API or MCP surface an external agent could drive on the public surface reviewed. Relevant to the agent ecosystem as a governance/review layer for agent output, not as a callable component.

Agent-friendly score: 5/10

Evidence

Open-source, self-hostable workspace for running coding agents — source (2026-09-01) verified
Agent-agnostic (Claude Code, Codex, Cursor, OpenCode, Grok) — source (2026-09-01) verified
Isolated git worktrees per task; workflows chain agent steps + human approvals — source (2026-09-01) verified
AGPL-3.0 licensed, source on GitHub — source (2026-09-01) verified

scorecard.json · transparency passport · registry · methodology

More: compare agents · best of · developer tools · incident registry

Verdict by Hlido Editor, our automated editorial system · Method: public-surface-tier-1+editorial-narrative-v2 · Methodology version 2026.05 ·

How this page was produced. The scores, claim verdicts and evidence come from automated hands-on testing of the product’s public surface. The written analysis is drafted by an AI system, and pages publish without a person reviewing each one. Hlido publishes this record and answers for it — tell us if anything here is wrong and we will correct it.

Embed this trust badge

Live, always-current independent score — free to embed on your site or README. No vendor pays for placement.

Markdown

[![Hlido trust score](https://hlido.eu/badge/proliferate-ai-proliferate.svg)](https://hlido.eu/check/?agent=proliferate-ai-proliferate)

HTML

<a href="https://hlido.eu/check/?agent=proliferate-ai-proliferate"><img src="https://hlido.eu/badge/proliferate-ai-proliferate.svg" alt="Hlido trust score"></a>