MCP简体中文Upgrade
Market/Categories/Developer & AI Platform/PMB

Independent product intelligence · Worldwide · 2026-05-01 → 2026-07-31

PMBpmbai.dev

Stores coding assistants’ project history locally.

Developer & AI Platformpmbai.devTracked since 2026-08-191 discovery source
553Backlinks
54Referring domains
17Domain authority

Authority & distribution

How strong is PMB’s web footprint?

Backlinks measure accumulated distribution; referring domains show how broadly that authority is spread. Both are observable on the open web, so both are published for every product here.

Backlinks

553Current snapshot

Referring domains

54Current snapshot

Authority

17Domain authority · category median 22
At 17, PMB sits below the 22 median for its category.

Trajectory · Pro

The counts above are the latest of 57 daily observations. The curve through them is on the full profile.

See the trajectoryOpen the full profile

Traffic & engagement

Where PMB ranks

Measured · Pro

Measured for this product: global rank.

Unlock this profileOpen the full profile

AI referral visibility

Which AI assistants are already sending PMB traffic?

Measured · Pro

Measured for this product: AI referral traffic.

Unlock this profileOpen the full profile

Product profile

What PMB is selling.

Traffic makes more sense when you can connect it to the actual product promise.

“PMB”

Stores coding assistants’ project history locally.

Developer & AI PlatformSeen 2026-08-19 → 2026-08-19Coverage tier C

What we have measured

4 of 8 signal groups are available for this product: identity, authority and backlinks, AI referrals, search demand.

Not measured for this domain: traffic and engagement, audience demographics, country distribution, acquisition mix. Those sections are left out of the page rather than filled with estimates.

Canonical website: https://pmbai.dev

Observed across 1 discovery source.

Research questions

What the evidence answers about PMB.

Every answer is tied to a source we collected, and the questions with no evidence behind them are left unanswered rather than guessed.

What problem does PMB solve for AI coding agents?

PMB solves the problem of AI coding agents forgetting project context between sessions by storing their history locally. It provides a persistent memory that is automatically injected before the agent thinks, eliminating the need for the agent to recall information manually.

What features and interfaces does PMB offer?

PMB offers a local-first memory system with features including automatic memory injection, sub-millisecond writes, hybrid recall with 94.5% recall@10, and an honest follow-rate scoring system. It includes a local web dashboard with a live entity graph (The Map) and a git-graph journal (The Timeline), and integrates as an MCP tool.

How is PMB priced or packaged?

PMB is offered as open-source software under the Apache-2.0 license. It is described as having no API keys, no cloud, and no telemetry, indicating a free-to-use, self-hosted model.

What is PMB used for and in what situations?

PMB is used to give AI coding agents a persistent, local-first memory of project decisions, lessons, and facts. It is used in situations where a developer works with coding assistants like Claude Code, Cursor, Codex, or Zed, ensuring the agent has consistent context across multiple interactions without cloud dependency.

Who is PMB designed for?

PMB is designed for developers and AI coding agent users. It is specifically built for individuals who use coding assistants and want to maintain project memory and context for their AI agents in a local, offline environment.

What is PMB?

PMB is a local-first memory tool for AI coding agents that stores project history, decisions, lessons, and facts in a single SQLite file on the user's disk. It is an open-source tool designed to give agents like Claude Code, Cursor, and Codex persistent memory without cloud services or API keys.

Behind the lock

Unlock full PMB intelligence.

  • Backlink trajectoryThe figures above are the latest observation. The daily curve behind them holds 57.
  • AI referral detailHow much traffic assistants send.
  • Search structure and ranking depthOrganic against paid, brand against non-brand, and how many keywords rank in the top three against the long tail — which is what decides whether the demand is portable or tied to the name.
  • The rest of the marketPage through every leaderboard 100 rows at a time, put any five products side by side, and query the whole dataset over API / MCP.

Worldwide estimates · Data period 2026-05-01 → 2026-07-31. Figures are measured estimates intended for market research. Signals we have not measured for this domain are omitted, never estimated; a measured zero is still reported as zero. How this was measured