MCP.Market

Colber MCP server

by Obi49·io.github.Obi49/colber-mcp·v0.1.1
No reviews yet

Trust + coordination infrastructure for the agent economy. 27 tools across 6 Colber modules.

B72/100grade B
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
B72/100

full report

Adoption
New

0 stars41 downloads/wk

Reviews

Write one

Nobody has reviewed Colber yet.

If you have run it, two minutes of your experience saves the next person an afternoon.

Colber tools (27, 4 write)

write = sends, deletes, buys or posts

Read from the package source without running it. The installed server may list more.

  • colber_identity_register

    [Colber] Register a new agent identity from an Ed25519 public key. Returns the derived did:key, the internal agent UUID, and the registration timestamp.

  • colber_identity_resolve

    [Colber] Resolve a DID to its agent record.

  • colber_identity_verify

    [Colber] Verify a signature against the public key bound to a DID. Returns { valid: true|false }.

  • colber_insurance_claim

    [Colber] File a claim against an existing policy. The claim stays in `open` until the admin endpoint decides (v1 has no oracle-driven arbitrator). Idempotent on `(policyId, idempotencyKey)`.

  • colber_insurance_quote

    [Colber] Compute a premium quote for a delivery deal. Risk multiplier is derived from the reputation score of `subscriberDid`. No commitment is made; nothing is persisted.

  • colber_insurance_status

    [Colber] Read the full state of a policy: policy fields + simulated escrow holding + all claims filed against it.

  • colber_insurance_subscribe

    [Colber] Create a policy and lock the simulated escrow. Re-quotes server-side so the client cannot inject a stale premium. Idempotent on `idempotencyKey`. Rejected if the global exposure cap would be exceeded.

  • colber_memory_retrieve

    [Colber] Semantic search across memories visible to the caller (top-k by cosine similarity). Permission-aware: only memories the caller is allowed to see can be returned.

  • colber_memory_share

    [Colber] Grant additional agents read access to a memory. Owner-only. Optional `expiresAt` is recorded for future expiry enforcement.

  • colber_memory_store

    [Colber] Persist a new memory (text + structured payload). Generates an embedding via the configured provider. Returns the new memory id.

  • colber_memory_updatewrite action

    [Colber] Update a memory's text and/or payload. Versioned (audit trail in memory_versions). Re-embeds when text changes. Owner-only.

  • colber_negotiation_counter

    [Colber] Submit a counter-proposal in response to an existing proposal. Strategy decides whether the counter replaces the prior proposal (multi-criteria) or competes as a fresh bid (ascending-auction).

  • colber_negotiation_propose

    [Colber] Submit a signed proposal (bid or scored offer) to a negotiation. Validates the Ed25519 signature against the supplied public key + the strategy rules. Returns the updated NegotiationView.

  • colber_negotiation_settle

    [Colber] Finalize the deal: verifies a signature from EVERY partyDid over `{negotiationId, winningProposalId}` (JCS canonical form). Records `negotiation.settled`. Idempotent on subsequent calls (returns the settled state).

  • colber_negotiation_start

    [Colber] Create a new negotiation with terms (subject, strategy, constraints, deadline, parties). Returns the freshly-created NegotiationView. Idempotent on `idempotencyKey`.

  • colber_observability_alert_createwrite action

    [Colber] Create a declarative alert rule. Storage only \u2014 the evaluation engine is not implemented in this service yet.

  • colber_observability_alert_deletewrite action

    [Colber] Delete an alert rule. Returns `{ deleted: true, id }`.

  • colber_observability_alert_get

    [Colber] Read a single alert rule by id.

  • colber_observability_alert_list

    [Colber] List alert rules owned by an operator.

  • colber_observability_alert_patchwrite action

    [Colber] Partially update an existing alert rule. Provide only the fields you want changed.

  • colber_observability_log

    [Colber] Ingest one or more log events. Batched for ClickHouse insertion. Returns the count of accepted events plus per-event rejection reasons for any that failed validation.

  • colber_observability_query

    [Colber] Structured search over logs or trace spans within a time range. Returns matching rows ordered by timestamp descending, paginated by limit + offset.

  • colber_observability_trace

    [Colber] Ingest one or more W3C-aligned trace spans. Batched for ClickHouse insertion. Returns the count of accepted spans plus per-span rejection reasons.

  • colber_reputation_feedback

    [Colber] Submit a signed feedback after a transaction. Validates the issuer's signature against their DID and enforces (feedbackId, [from,to,tx]) idempotency.

  • colber_reputation_history

    [Colber] Paginated history of the agent's transactions and feedbacks. Cursor-based: pass `nextCursor` from the previous page on the next call.

  • colber_reputation_score

    [Colber] Return the agent's signed reputation score envelope (did, score 0..1000, scoreVersion, computedAt, ed25519 attestation).

  • colber_reputation_verify

    [Colber] Verify a signed reputation attestation against the platform public key. Crypto-only; no DB lookup.

Public scan report

scanner v0.1.9 · 2026-09-27 · same rubric, same numbers if you re-run it

no findings
  • Code scan2 source files scanned25/25
  • –Live reliabilityno gateway calls yet and no remote to proben/a
  • –Tool poisoningtools not inspected (local package is not executed); not countedn/a
  • Auth qualitystatic API keys via environment variables6/15
  • Maintenancelast push 142 days ago8/15
  • Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10
Overall 72/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Colber compares

Against every graded AI and agents MCP server, from our own scans and uptime checks. Updated after each scan run.

Safety score
72/100
higher than 40% of 4,460 AI and agents servers
Tools
27
the typical AI and agents server has 6
Tools that change things
15%
typical: 0%. Fewer write tools means less an agent can break
Downloads a week
41
more than 33% of 1,538 AI and agents servers with a package

What its tools let an agent do: 23 read · 4 change

Install Colber in Claude Code, Cursor or VS Code

Runs npx -y @colber/mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add colber-mcp -- npx -y @colber/mcp
Add to Cursor

What the publisher says

From the Colber repository's README, as published. We do not edit it. Read it on GitHub

Colber

Trust, coordination & continuity — for the agent economy.

Five integrated services AI agents need to operate at scale: Reputation · Memory · Observability · Negotiation · Insurance — exposed via SDKs (TypeScript, Python) and the Model Context Protocol.

🌐 * · 📦 npm @colber/ · 🐍 PyPI colber-sdk · 🔌 MCP-native

What is Colber?

Colber is the infrastructure layer of trust, coordination and continuity for the agent economy. The hosted platform at exposes five integrated capabilities through one consistent surface (REST · gRPC · MCP):

Plus an identity support service (DID:key + Ed25519 signature verification) used by every module.

🏗️ Status — v1 shipped. All five modules + identity are live on https://colber.dev, end-to-end tested (23/23 green), with 27 MCP tools published.

What's in this repo (open core)

This repository is the public open-core of Colber. It contains the code you need to integrate with Colber from your own agent, plus the public protocol contract:

apps/
├── sdk-typescript/   →  npm  @colber/sdk
├── sdk-python/       →  PyPI colber-sdk
├── mcp-server/       →  npm  @colber/mcp   (CLI: npx -y @colber/mcp)
└── site/             →  https://colber.dev (landing source)

packages/
├── core-types/       →  Public protocol types (errors, envelopes, DIDs)
├── core-crypto/      →  Client-side crypto helpers (DID:key, Ed25519, JCS canonicalisation)
├── core-config/      →  Env-var validation utilities (zod schemas)
└── core-logger/      →  Structured logging utilities (pino + traceId)

tooling/              →  Shared TS / ESLint configs
.github/              →  Issue + PR templates
docs/diagrams/        →  High-level architecture diagrams (Mermaid)
docs/MCP_REGISTRIES.md →  Submission templates for Anthropic, Smithery, mcp.so

Everything in this repo is Apache-2.0. You can fork it, embed it in your products, ship modified versions of the SDKs, contribute back via PR.

Not in this repo

The server-side implementation of the five modules + identity (the actual Reputation engine, Memory vector index, Observability ingestion, Negotiation event store, Insurance escrow logic, operator console) is proprietary and runs on https://colber.dev. To use it, you call the hosted endpoints from the SDKs or the MCP server. This is the standard open-core model used by Stripe, Datadog, Auth0.

Quick start

TypeScript

npm install @colber/sdk
import { ColberClient } from '@colber/sdk';

Shortened. The full README is on GitHub.

Nothing above is checked by us. What we check is on the safety report.

Colber: common questions

Is Colber MCP server safe?
Mostly: it is graded B (72/100). Read the Colber safety report
How do I install Colber?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Colber need an API key?
Yes. The registry entry asks for COLBER_AUTH_TOKEN.
Is Colber maintained?
The last commit was 179 days ago (2026-05-08). The latest release is v0.1.1.
What can I use instead of Colber?
Servers from other publishers that do the same job: LLM Sandbox MCP server, Agent Lsp MCP server and Grandet MCP server.
How does Colber compare with other AI and agents MCP servers?
safety score 72/100 (higher than 40% of 4,460 AI and agents servers); tools 27 (the typical AI and agents server has 6); tools that change things 15% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Colber

Same job from other publishers: the closest match first, then the best rated.

  • LLM Sandbox
    Securely run LLM-generated code in isolated containers across 7 languages and 3 container backends.
    A
  • Agent Lsp
    Orchestrates language servers into 65 code-intelligence tools across 32 languages, token-optimized.
    A
  • Grandet
    Cheapest API for any model across official, inference and relay providers; find free credits
    A
  • three.ws Agora
    Agora agent economy over MCP: browse the board, register, claim on-chain work, and post bounties.
    B
  • Skillsmith
    Share agent skills across your team, scan them for risk, and track what's actually used.
    B