Zaxy MCP server

by syndicalt·io.github.syndicalt/zaxy·v3.2.0·17 stars

No reviews yet

Event-sourced, auditable agent memory over MCP: cited checkout, replayable provenance.

A88/100grade A

Write a review Install

What users say

No reviews yet

Be the first

Safety scan

A88/100

full report

Adoption

Growing

17 stars

Reviews

Write one

Nobody has reviewed Zaxy yet.

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

Zaxy tools (39, 4 write)

write = sends, deletes, buys or posts

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

context_after_turn
Persist a completed turn and return compact context for the next turn.
coordination_apply_approval
Apply remote approval decisions to a coordination mission.
coordination_approval_packet
Return a portable pending/conflicted finding packet for a remote reviewer.
coordination_assign
Assign scoped work to a coordination worker.
coordination_checkout
Return accepted coordination state for prompt injection, with optional diagnostics for pending or conflicted findings.
coordination_handoffwrite action
Create a final coordination handoff event for a mission.
coordination_merge_brief
Return a replay-backed coordination brief for a mission.
coordination_performance_ledger
Return worker-level coordination outcome metrics for a mission.
coordination_promote
Promote an accepted finding into the parent mission history.
coordination_proof_trace
Replay a Coordinate proof packet chain by artifact_id, handoff_id, or proof_seq.
coordination_report_finding
Record a worker-local coordination finding with evidence; it is not trusted parent state until reviewed and promoted.
coordination_review_finding
Review a worker finding as accepted, rejected, deferred, or conflicted.
coordination_start
Start a parent coordination mission session.
coordination_worker_createwrite action
Register a worker session under a parent mission.
fleet_audit
Return full replay-only provenance for every fleet memory: origin actor/session, source + gate citations.
fleet_review
Steward review of a held fleet promotion (accepted activates a pending memory).
fleet_status
Return a replay-backed fleet brief: active and pending promotions, enrolled agents, and trust tiers.
memory_bootstrap
At session start, return compact Zaxy memory guidance and the recommended first checkout call.
memory_causal_predecessors
Read directed causal causes of an entity from graph-backed memory.
memory_causal_successors
Read directed causal effects of an entity from graph-backed memory.
memory_claim_confidence
Score claim confidence from cited support and conflict evidence.
memory_confidence_trajectory
List append-only confidence assessments for a claim.
memory_consolidation_candidate
Append a cited, review-pending consolidation candidate event.
memory_consolidation_propose_from_logwrite action
Create non-authoritative consolidation candidates from Eventloom log segments.
memory_consolidation_review
Append a consolidation candidate review event without promoting authority.
memory_consolidation_status
Read review-gated consolidation candidate status.
memory_explain_outcome
Explain an outcome from cited causal and checkout memory context.
memory_known_unknowns
List replay-derived known unknowns for a session.
memory_plan_from_procedures
Build a non-authoritative planning packet from applicable procedures.
memory_propose_belief_updatewrite action
Append a cited, review-pending belief update proposal without promoting authority.
memory_record_known_unknown
Record a cited, open, non-authoritative known unknown.
memory_replay
Replay events from a session starting at a sequence number.
memory_reverification_needs
List open unknowns, unresolved conflicts, and low-confidence claims needing re-verification.
memory_similar_procedures
Retrieve similar procedure candidates from Skill Memory and consolidation memory.
memory_skill
Append a typed skill lifecycle event and project it into Skill Memory.
memory_synthesis_artifact
Persist Memory Checkout answer candidates as synthesis artifacts and optional outcome feedback.
memory_synthesis_evidence
Record feedback for one Memory Checkout synthesis ledger row.
memory_verbatim
Retrieve exact Eventloom source chunks with citations.
subagent_cleanup
Finalize a subagent session and return its handoff bundle.

Public scan report

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

no findings

Code scan191 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 47 days ago12/15
Maintainer identityregistry namespace matches repository owner; GitHub account older than a year8/10

Overall 88/100. Components that don't apply are left out of the denominator. Any critical finding is an F.Rubric Appeal a finding JSON

How Zaxy compares

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

Safety score: 88/100; higher than 80% of 4,460 AI and agents servers
Tools: 39; the typical AI and agents server has 6
Tools that change things: 10%; typical: 0%. Fewer write tools means less an agent can break
GitHub stars: 17; more than 88% of 2,711 AI and agents servers

What its tools let an agent do: 35 read · 4 change

Install Zaxy in Claude Code, Cursor or VS Code

claude mcp add zaxy -- uvx zaxy-memory

Add to Cursor

What the publisher says

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

Zaxy

Production memory for agent teams that need receipts.

Zaxy turns agent context into an auditable project memory fabric. It captures parent missions, worker sessions, tool observations, cited findings, conflict review, approval packets, and accepted merge-back into one durable history that can be queried, replayed, and inspected.

Under the hood, Zaxy uses Eventloom append-only JSONL as the source of truth and an embedded LadybugDB graph projection for local reasoning. It is built for agents that need to remember what happened, cite where it came from, and avoid turning project state into a pile of markdown files and vector chunks.

The embedded LadybugDB graph projection is the default local runtime.

The plain install uses embedded LadybugDB. Install zaxy-memory[neo4j] only for the optional Neo4j sidecar, and zaxy-memory[pathlight] only for Pathlight tracing.

Why It Matters

findings are reviewed and merged.

Auditable memory: every accepted fact can point back to Eventloom history.
Agent-team coordination: parent and worker sessions stay isolated until

LangGraph, CrewAI, and AutoGen can use the same memory interface.

Local-first runtime: the default path uses embedded LadybugDB, no Neo4j sidecar.
MCP-native integration: Codex, Claude Code, Cursor, VS Code, Hermes Agent,

Benchmark claims withdrawn (2026-07-03). The prior LongMemEval numbers

were withdrawn: they were produced in oracle mode (mean ~1.9 candidate

sessions per question, so Recall@5/citation-coverage were ~1.0 by

construction, not by retrieval) and the preference-question scores rested on a

hardcoded answer table that has since been removed. Zaxy does not currently

publish a LongMemEval score. A real, full-haystack LongMemEval run is planned;

until it lands, treat the earlier 0.956/0.910/1.000 figures as retracted.

The Harvey LAB claim is pending the same audit.

Quick Start

Install, init, verify

pipx install zaxy-memory
zaxy init
zaxy memory log --eventloom-path .eventloom --limit 5
zaxy memory bootstrap --eventloom-path .eventloom
zaxy doctor --eventloom-path .eventloom

The PyPI distribution is zaxy-memory; the import package and console command are still zaxy. Bare zaxy init sets up the local embedded graph posture, repo-local profile, deterministic capture config, genesis event, heartbeat, and MCP guidance. For Codex, the printed activation launcher starts the managed capture watcher when the local capture config is present; pass --capture start only when you want init itself to start the watcher before opening Codex. The default human output is compact and action-first; add --verbose when you need the full setup diagnostics, optional checks, fallback commands, resume guidance, and notes. For automation, zaxy init --json keeps the raw onboarding fields and adds setup.status, setup.issues, setup.pending, readiness.status, readiness.reasons, readiness.actions, and structured readiness.actionitems for both commands and non-command review tasks. Each structured action carries label, command, original source, and hints for compact-output tips such as activation replacement and path-stable command guidance. Installers can render those tips without parsing prose. It also includes setup.summary, readiness.summary, readiness.requiredactioncount, and readiness.reasoncount, so client UIs can render compact status without parsing human output. It also separates readiness.blockingdiagnostics from readiness.nonblocking_diagnostics so scripts can distinguish setup completion, required actions, and advisory doctor warnings before relying on live memory.

Shortened. The full README is on GitHub.

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

Zaxy: common questions

Is Zaxy MCP server safe?: Yes, by our scan: it is graded A (88/100). Read the Zaxy safety report
How do I install Zaxy?: It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Zaxy need an API key?: Not as far as the registry entry and our scan can tell: no credentials are declared or required.
Is Zaxy maintained?: The last commit was 86 days ago (2026-08-10). The latest release is v3.2.0.
What can I use instead of Zaxy?: Servers from other publishers that do the same job: CHAP Coordinator MCP server, agent-device MCP server and Basic Memory MCP server.
How does Zaxy compare with other AI and agents MCP servers?: safety score 88/100 (higher than 80% of 4,460 AI and agents servers); tools 39 (the typical AI and agents server has 6); tools that change things 10% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.

Alternatives to Zaxy

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

CHAP Coordinator
Auditable records of human decisions over AI agent work. Approvals, edits, overrides, escalations.
not reviewed
Established
A
agent-device
Open source device layer for AI agents: control mobile, web, TV, and desktop apps over MCP
not reviewed
Widely used
A
Basic Memory
Local-first knowledge management with bi-directional LLM sync via Markdown files.
not reviewed
Widely used
B
Pm Claude Skills
1310 professional Agent Skills + workflow recipes — searchable & fetchable over MCP.
not reviewed
Established
A
IWE (iwe.md)
Markdown knowledge base as agent memory. Runs against the notes directory it is started in.
not reviewed
Established
A