# Bluet (npm · bluet-mcp)

Product memory your agents read before they work: briefs, plan checks against decisions, debriefs.

- Trust score: 63/100 (medium)
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-09-26

## Components

- remote · `bluet.dev`: 39/100, [markdown](https://verifymcp.io/servers/dev-bluet-bluet/bluet.md), [page](https://verifymcp.io/servers/dev-bluet-bluet/bluet)
- npm · `bluet-mcp`: 63/100 (this document), [markdown](https://verifymcp.io/servers/dev-bluet-bluet/bluet-mcp.md), [page](https://verifymcp.io/servers/dev-bluet-bluet/bluet-mcp)

## Channel facts

- Registry: `npm`
- Package: `bluet-mcp`
- Version: `0.5.1`
- Transport: `stdio`

## Trust breakdown

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-09-26.

- **Supply Chain Security**: 98/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 38 of 119 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 19/100
  - Repository check failed: no source repository is declared.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (Apache-2.0).
  - Actively maintained (last published 1 days ago).
  - Security-disclosure policy not yet verified: we couldn't inspect the source repository.
- **Schema Quality & AI Usability**: 73/100
  - AI-judged instruction clarity (good).
  - Tool/resource definitions use about 2290 tokens (~76/item across 30 items; 30 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 7/100
  - Stability observed for 2 of 30 days with no destabilising changes; credit accrues until the full window elapses.
- **Tool Coverage**: 75/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 26% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - We read all 30 captured tool definition(s), and no name or description among them implies an irreversible operation.
  - An AI judge read all 31 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

## Install

### How do I install the Bluet MCP server?

Bluet runs locally as an npm package, launched with npx -y bluet-mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

### Claude

```bash
claude mcp add dev-bluet-bluet -- npx -y bluet-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "dev-bluet-bluet": {
      "command": "npx",
      "args": [
        "-y",
        "bluet-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "dev-bluet-bluet": {
      "command": "npx",
      "args": [
        "-y",
        "bluet-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add dev-bluet-bluet -- npx -y bluet-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "dev-bluet-bluet": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "bluet-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add dev-bluet-bluet --command npx --arg -y --arg bluet-mcp
```

### Hermes

```yaml
mcp_servers:
  dev-bluet-bluet:
    command: "npx"
    args: ["-y", "bluet-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "dev-bluet-bluet": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "bluet-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add dev-bluet-bluet -t stdio -c npx -a -y bluet-mcp
```

### Other

```json
{
  "mcpServers": {
    "dev-bluet-bluet": {
      "command": "npx",
      "args": [
        "-y",
        "bluet-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-09-26 (score 63, +40)

- [security improvement] Injection markers: unverified → pass
- [security improvement] Known CVEs: unverified → pass
- [functional improvement] MCP protocol: unverified → pass
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] Stability: unverified → 0.07
- [functional improvement] Dependency health: unverified → 0.86
- [functional] Package version: 0.4.9 → 0.5.1

### 2026-09-25 (score 23, −29)

- [security regression] Tool safety: pass → unverified
- [security improvement] Known CVEs: unverified → pass
- [security] Stability: Stability not yet verified: we do not have a sandbox capture of the MCP schema this version of the package serves yet.
- [functional regression] Capabilities: pass → unverified
- [functional regression] Tool coverage: 100 → unverified
- [functional improvement] Dependency health: unverified → 0.86
- [functional] First check of Schema quality: unverified
- [functional] Package version: 0.4.9 → 0.5.1
- [functional] Package version: 0.4.9 → 0.5.0
- [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server

### 2026-09-24 (score 52)

First indexed and scored.

## MCP tools (30)

### `bluet_products_list` (~18 tokens)

List products

List all products in Bluet memory

### `bluet_context_get` (~49 tokens)

Get a product's full context

Get full product context (strategy, bets, roadmap, releases, decisions, gotchas, briefs, constraints)

Input parameters:

- `product` (string, required): Product slug or id (e.g. pasiv)

### `bluet_decision_list` (~34 tokens)

List decisions

List decisions for a product

Input parameters:

- `include_superseded` (boolean)
- `product` (string, required)

### `bluet_decision_get` (~25 tokens)

Get a decision

Get a single decision by id

Input parameters:

- `id` (string, required)

### `bluet_gotcha_list` (~26 tokens)

List rules

List gotchas for a product

Input parameters:

- `product` (string, required)

### `bluet_gotcha_search` (~33 tokens)

Search rules

Search gotchas by text query

Input parameters:

- `product` (string, required)
- `query` (string, required)

### `bluet_drift_list` (~45 tokens)

List conflicts

List drift: where product memory and the outside world disagree. Each item carries the tool call that resolves it, or why none exists.

Input parameters:

- `product` (string, required)

### `bluet_roadmap_list` (~26 tokens)

List roadmap items

List roadmap items for a product

Input parameters:

- `product` (string, required)

### `bluet_release_list` (~23 tokens)

List releases

List releases for a product

Input parameters:

- `product` (string, required)

### `bluet_brief_get` (~39 tokens)

Get an agent's brief

Get an agent brief (or all briefs if agent_key omitted)

Input parameters:

- `agent_key` (string)
- `product` (string, required)

### `bluet_constraint_list` (~32 tokens)

List constraints

List derived constraints from strategy non-goals/principles and agent briefs

Input parameters:

- `product` (string, required)

### `bluet_decision_record` (~99 tokens)

Record a decision

Record a product decision (optionally superseding a prior one)

Input parameters:

- `bet` (string): The bet this decision serves, by name or id. Counted as evidence on Intent
- `decision` (string, required)
- `product` (string, required)
- `rationale` (string)
- `source` (string)
- `status` (string)
- `supersedes_id` (string)
- `title` (string, required)

### `bluet_decision_update` (~60 tokens)

Update a decision

Edit fields on an existing product decision without creating a new version

Input parameters:

- `decision` (string)
- `id` (string, required)
- `rationale` (string|null)
- `status` (string)
- `title` (string)

### `bluet_product_create` (~71 tokens)

Create a product

Create a product to hold memory. Required before anything else on a fresh install.

Input parameters:

- `description` (string)
- `name` (string, required): Display name, e.g. Pasiv
- `slug` (string): URL-safe id; derived from name when omitted
- `url` (string)

### `bluet_strategy_set` (~204 tokens)

Set a product's strategy

Set the product strategy. Supersedes the current one, keeping the previous version linked rather than overwriting it.

Input parameters:

- `measure_comparison` (string): What it was before, e.g. 'from 41% in June'
- `measure_detail` (string): What the measure is and why it is the one that counts
- `measure_label` (string): Name of the single measure, e.g. 'The one measure'
- `measure_value` (string): Its current value, e.g. '68%'
- `non_goals` (string): One per line; becomes a constraint
- `north_star` (string): The one sentence the product exists to make true. Shown as the Intent headline
- `positioning` (string)
- `principles` (string): One per line; becomes a constraint
- `product` (string, required)
- `source` (string)
- `summary` (string)
- `title` (string, required)

### `bluet_bet_list` (~31 tokens)

List bets

List bets for a product

Input parameters:

- `product` (string, required)
- `status` (string)

### `bluet_bet_upsert` (~172 tokens)

Record or update a bet

Create or update a bet — a hypothesis the product is testing

Input parameters:

- `conviction` (string): How far it has been borne out. Distinct from status, which is lifecycle
- `falsifier` (string): What would change our mind. A bet with no falsifier cannot be lost, only defended
- `hypothesis` (string): The claim, in one sentence
- `id` (string)
- `priority` (number): Higher sorts earlier; this is the order of the argument
- `product` (string, required)
- `reasoning` (string): Why we believe it — the argument, not the claim
- `started_at` (string): ISO date the bet was actually taken, when that is not today
- `status` (string)
- `title` (string, required)

### `bluet_gotcha_add` (~85 tokens)

Record a rule

Add a gotcha / trap / tribal knowledge note

Input parameters:

- `bet` (string): The bet this gotcha is evidence about, by name or id
- `body` (string, required)
- `product` (string, required)
- `severity` (string)
- `source` (string)
- `tags` (string)
- `title` (string, required)

### `bluet_roadmap_upsert` (~70 tokens)

Record or update a roadmap item

Create or update a roadmap item

Input parameters:

- `description` (string)
- `horizon` (string)
- `id` (string)
- `product` (string, required)
- `sort_order` (number)
- `status` (string)
- `title` (string, required)

### `bluet_release_upsert` (~104 tokens)

Record or update a release

Create or update a release plan

Input parameters:

- `bet_id` (string|null): Link to the bet this delivers
- `decision_id` (string|null): Link to the decision that authorised it
- `id` (string)
- `name` (string, required)
- `notes` (string)
- `platforms`
- `product` (string, required)
- `status` (string)
- `target_date` (string)
- `version` (string)

### `bluet_brief_set` (~61 tokens)

Set an agent's brief

Create or update an agent brief

Input parameters:

- `agent_key` (string, required)
- `brief` (string, required)
- `constraints` (string)
- `goals` (string)
- `product` (string, required)
- `title` (string, required)

### `bluet_brief_push` (~78 tokens)

Send agent briefs to a webhook

Push agent brief(s) to endpoints (webhook) and/or queue for MCP poll. Always upserts the canonical brief store and records a delivery row.

Input parameters:

- `agent_name` (string): Single agent key; omit with all=true
- `all` (boolean): Push every brief for the product
- `product` (string, required)

### `bluet_ingest` (~91 tokens)

Ingest a note or link

Ingest a note/url/payload into Bluet as decision|gotcha|roadmap|release|brief|signal

Input parameters:

- `notes` (string)
- `payload` (object)
- `product` (string, required)
- `title` (string)
- `type` (string, required): decision | gotcha | roadmap | release | brief | signal | note
- `url` (string)

### `bluet_preflight` (~114 tokens)

Brief an agent before a task

CALL FIRST, before changing code. Given a task (and optionally the files you expect to touch), returns what you must know: rules not to break, decisions that constrain the work, non-goals, the bet it serves, and open questions — each numbered and sourced. One call replaces reading every table.

Input parameters:

- `files` (array): Paths you expect to touch
- `product` (string, required): Product slug or id
- `task` (string, required): What you are about to do, in a sentence or two

### `bluet_check` (~81 tokens)

Check a plan against decisions

Check a plan or diff against everything the product has decided. Returns conflicts, each citing the decision, rule or non-goal it collides with. verdict 'conflicts' with severity 'blocking' means stop and ask the user.

Input parameters:

- `plan` (string, required): The plan, or a diff / summary of the change
- `product` (string, required)

### `bluet_debrief` (~110 tokens)

Propose what a session settled

Call when you finish work that settled something. List the decisions you made (a choice between options, a constraint you adopted) and traps you hit. They are PROPOSED for the owner to accept, never recorded directly. The summary is also checked against the record, and a contradiction is raised as drift.

Input parameters:

- `decisions` (array)
- `gotchas` (array)
- `product` (string, required)
- `summary` (string, required): What the session did, in a few sentences

### `bluet_brief_today` (~84 tokens)

Today's product brief

The last day for a product: what is waiting on the owner (proposals, drift), recent activity, the rules agents are told, and questions only the owner can answer.

Input parameters:

- `format` (string): 'session' returns the short text a session starts with, without the written brief; 'status' returns counts only
- `product` (string, required)

### `bluet_proposals_list` (~68 tokens)

List what awaits the owner

List what is waiting for the owner's judgment: decisions read from pull requests and issues, traps, drift, and what agents debriefed. Show these to the user; accept or dismiss only on their say-so with bluet_proposal_judge.

Input parameters:

- `product` (string, required)

### `bluet_proposal_judge` (~49 tokens)

Accept or dismiss a proposal

Accept or dismiss one pending proposal — ONLY when the user has said to. Accepting writes it into the record.

Input parameters:

- `action` (string, required)
- `id` (string, required)

### `bluet_repo_ingest` (~150 tokens)

Record a repository read locally

Record a repository's material read on the caller's machine (README, files written for agents, merged pull requests, issues) against a product. Used by `npx bluet-mcp init` when linked to bluet.dev, so a repository GitHub will not let bluet.dev read can still be read. Same rule as a sync: transcriptions are recorded, inferences are proposed.

Input parameters:

- `description` (string|null)
- `files` (array)
- `issues` (array)
- `product` (string, required)
- `pulls` (array)
- `readme` (string|null)
- `repo` (string, required): owner/name, or the directory name

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/dev-bluet-bluet/bluet-mcp#diagnostics

## Score history

- 2026-09-26: 63
- 2026-09-25: 23
- 2026-09-24: 52

## Common questions

### What is the Bluet MCP server?

Bluet is an MCP server listed in the public MCP registry as dev.bluet/bluet. Product memory your agents read before they work: briefs, plan checks against decisions, debriefs. This page covers its npm package (bluet-mcp).

### Is the Bluet MCP server safe to use?

Bluet scores 63 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 26 September 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

### What tools does the Bluet MCP server expose?

Bluet exposes 30 tools: bluet_products_list, bluet_context_get, bluet_decision_list, bluet_decision_get, bluet_gotcha_list, and 25 more. Their descriptions and schemas cost roughly 2,132 tokens of context every time the server is loaded.

### Is the Bluet MCP server still maintained?

Bluet is still listed as active in the MCP registry. We last reached this channel on 26 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the Bluet MCP server under?

Bluet declares the Apache-2.0 licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/bluet-mcp
- Socket report: https://socket.dev/npm/package/bluet-mcp
- Website: https://bluet.dev/
- Changelog RSS feed: https://verifymcp.io/servers/dev-bluet-bluet/bluet-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/dev-bluet-bluet/bluet-mcp.json
- HTML version of this page: https://verifymcp.io/servers/dev-bluet-bluet/bluet-mcp