MCP.Market

Octopus Blackboard MCP server

by octoryn·io.github.octoryn/octopus-blackboard·v0.3.1·7 stars
No reviews yet

Shared memory & coordination for AI coding agents — a blackboard, not an orchestrator.

A86/100grade A
Write a reviewInstall
What users say
No reviews yet
Be the first
Safety scan
A86/100

full report

Adoption
Growing

7 stars33 downloads/wk

Reviews

Write one

Nobody has reviewed Octopus Blackboard yet.

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

Octopus Blackboard tools (38, 1 write)

write = sends, deletes, buys or posts

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

  • board_assign

    Assign a task to an agent and notify them — records the assignee AND leaves a 'please look at task #N' message in that agent's inbox. Passive: the board records the ask; the agent reads its inbox and decides to act. Never launches the agent.

  • board_attribute

    Record a single attribution (commit + optional file) to a human or AI actor.

  • board_blame

    Blame → narrative: trace a file line back to the session that introduced it and surface that session's other work, decisions, and handoffs.

  • board_check

    Governance gate: assert policy over commits and report pass/fail plus violations. Read-only — reports, does not block. Use in CI to keep unreviewed AI work off a branch.

  • board_claim

    Claim a task by key so other agents know you own it. If another agent already holds it, the response reports the conflict (the board records but does not block).

  • board_decision

    Record a decision and its rationale so other agents see what was decided and why.

  • board_evidence

    Attach evidence (file path, URL, log, test run) supporting some work.

  • board_explain

    Explain a commit: its AI attribution, reviews, and related decisions.

  • board_export

    Export a portable attribution bundle (attributions, reviews, sessions, decisions) for a commit range, so attribution survives push/PR into a team board or CI.

  • board_file_changed

    Record that you touched a file. If a task key is given, the response reports files other agents already changed for that task (conflict awareness).

  • board_handoff

    Hand off work to another agent with a summary of the state at handoff.

  • board_handoffs

    Handoffs left FOR you — the work another agent passed to you, with context and open questions.

  • board_heartbeat

    Stamp the active session as alive (real-time liveness, so other agents can tell active from stale).

  • board_import

    Import an attribution bundle (as produced by board_export) into this board. Idempotent.

  • board_inbox

    Read what's waiting for you: messages (addressed or broadcast) and handoffs left for you.

  • board_ingest

    Ingest a CLI transcript's content into the active session (file edits, decisions, notes). Format is one of:

  • board_link

    Attribute a Git commit's files to the current session (who actually produced the code). Reads Git; never rewrites history. Optionally writes an additive git note.

  • board_message

    Leave a message for another agent (or broadcast with to='all').

  • board_note

    Broadcast a free-form status note to the board.

  • board_progress

    Report task progress 0–100 (moves it to in-progress, or done at 100). Use this from an agent as it works a task so the kanban shows a live progress bar.

  • board_prune

    Retention: delete messages/evidence/file-change rows created before an ISO time. The audit timeline is never pruned.

  • board_redact

    Hide a timeline entry's content at the read layer (the hash chain stays valid — this is not cryptographic erasure).

  • board_report

    Accountability scorecard: review coverage, AI/human ratio, per-agent breakdown, session and risk counts.

  • board_review

    Record who reviewed a commit and the outcome (human or AI reviewer).

  • board_risk

    Flag an open risk the next agent should know before acting.

  • board_sign

    Sign the current timeline head with the active session's key (attests board state through the head seq).

  • board_since

    Subscribe primitive: poll for timeline events after a given seq. Returns the new head seq and events (optionally filtered to those relevant to an agent). Call board_status first to get a starting seq.

  • board_status

    Read the board: active agents, open tasks and who holds them, unread messages, open risks, recent history. Call this before starting work to see what other agents are doing.

  • board_task

    Get a task's full kanban card (status, progress, assignees, active-agent count, impacted files, linked risks) by key or number (#145).

  • board_task_definewrite action

    Create or update a task's kanban fields: title, description (what it is), project, impact (change surface), and risk level. New keys get a stable task number.

  • board_tasks

    List all tasks as kanban cards (for a board/kanban view).

  • board_timeline

    Read the append-only, hash-chained history of everything that happened on the board.

  • board_trailers

    Git trailer lines encoding a commit's attribution, for embedding in a commit message.

  • board_trust

    Show signature trust: which sessions have signed the timeline and whether each signature is valid and still current.

  • board_unreviewed

    List AI-produced commits that have never been reviewed by a human.

  • board_who

    Who changed a file: Git authors, AI sessions that touched it, and commit attributions. Pass a line number to trace which session introduced that line.

  • session_start

    Start an execution session. Subsequent writes and links attribute to it. Captures machine, working directory, git branch, and repository.

  • session_stop

    Close the active session (or a specific one).

Public scan report

scanner v0.1.9 · 2026-09-26 · same rubric, same numbers if you re-run it

no findings
  • Code scan29 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 qualitylocal package, no credentials required12/15
  • Maintenancelast push 73 days ago12/15
  • Maintainer identityregistry namespace matches repository owner7/10
Overall 86/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Octopus Blackboard compares

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

Safety score
86/100
higher than 72% of 48,255 servers
Tools
38
the typical server has 7
Tools that change things
3%
typical: 0%. Fewer write tools means less an agent can break
GitHub stars
7
more than 86% of 28,688 servers
Downloads a week
33
more than 22% of 12,191 servers with a package

What its tools let an agent do: 37 read · 1 change

Install Octopus Blackboard in Claude Code, Cursor or VS Code

Runs npx -y octopus-blackboard on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add octopus-blackboard -- npx -y octopus-blackboard
Add to Cursor

What the publisher says

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

Octopus Blackboard

A shared memory and coordination layer for AI coding agents.

Agents do not need another boss. They need a shared blackboard.

简体中文

Part of Octopus Core — the open infrastructure stack for governed AI. One job per repo, along the agent lifecycle: Scout · Observe · Experience · Blackboard · Workstate · Runtime · Replay — with Inspect governing every stage. The whole stack rides one root primitive — Evidence, the canonical, tamper-evident atom that is the root category everything is built on.

This repo — Blackboard · Coordinate: Shared cognition for coding agents.

Claude Code, Codex, Gemini CLI, Cursor, and your local agents all work on the same repo — but they are blind to each other. One refactors auth while another rewrites the same file. Context is scattered across CLIs. Nobody can replay what happened.

Octopus Blackboard is not an orchestrator. It does not schedule agents, trigger them, or decide anything on their behalf. It is a passive, local-first shared memory that answers six questions:

Who is working on what      →  agents, tasks, claims
What changed                →  files_changed
What was decided            →  decisions
What evidence exists        →  evidence
What risks are open         →  risks
What was left for whom       →  messages, handoffs

Every agent only needs to be able to read the board, write the board, leave messages, and attach evidence. That's the whole contract.

Why

Enterprises are not primarily afraid that agents aren't smart enough. They are afraid that:

  • Multiple AI tools edit the same code simultaneously
  • Context is scattered across different CLIs
  • There is no shared memory, no audit, no handoff
  • There is no conflict awareness and no way to replay

The blackboard cuts straight through this. Every write appends a tamper-evident, hash-chained entry to an append-only timeline, so the entire history is auditable and replayable — and any after-the-fact edit to an earlier entry breaks verification.

Get Started in 2 Minutes

Blackboard is the easiest way into the Octopus stack — pure coordination memory that works today with just Claude Code and Cursor, no orchestrator required. One command sets everything up:

npx octopus-blackboard quickstart

That single command: creates the local board (.octoboard/) if it isn't there, auto-detects your MCP client (.claude/ → Claude Code, .cursor/ → Cursor, … otherwise a generic snippet), prints the exact paste-ready config for it, and proves the board works with one write + one read.

Then just:

board_note "hello" — that's your first board action, on a shared, tamper-evident timeline.

  1. Paste the printed config block into the file it names.
  2. Reload your MCP client (restart it, or reopen the project).
  3. You're done. Ask your agent to call board_status, then

Shortened. The full README is on GitHub.

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

Octopus Blackboard: common questions

Is Octopus Blackboard MCP server safe?
Yes, by our scan: it is graded A (86/100). Read the Octopus Blackboard safety report
How do I install Octopus Blackboard?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Octopus Blackboard need an API key?
No secret keys are declared. It reads 2 settings from the environment.
Is Octopus Blackboard maintained?
The last commit was 112 days ago (2026-07-15). The latest release is v0.3.1.
What can I use instead of Octopus Blackboard?
Servers from other publishers that do the same job: Deja Vu MCP server, Concord MCP server and Claudexor MCP server. Compare all Octopus Blackboard alternatives.
How does Octopus Blackboard compare with other MCP servers?
safety score 86/100 (higher than 72% of 48,255 servers); tools 38 (the typical server has 7); tools that change things 3% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Octopus Blackboard

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

All Octopus Blackboard alternatives →
  • Deja Vu
    deja-vu: local memory over the session histories of forty-three coding agents.
    A
  • Concord MCP
    Cross-harness communication and shared work-state for AI coding agents.
    A
  • Claudexor
    Route coding work across Claude Code, Codex, Cursor, and OpenCode with shared context.
    A