MCP Server for Z.AI - A Model Context Protocol server that provides AI capabilities
Nothing to run from here
AXIOM is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
94/100
trust score
44
downloads a day
local
runs on your machine
2k
downloads in 30d
What it does
Transactional write gate for coding agents: plan, canonical manifest, checks, two-phase apply.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user dragoscv-axiom-mcp '--' npx -y @codai/axiom-mcp mcp --root '<root>'
# 2 — ask your agent for something these tools do
# axiom_plan_validate, axiom_plan_compile, axiom_manifest_verify and 16 more
Put your own value in place of <root> before you run this.
AXIOM costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
DRAll 1 server
dragoscv1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts AXIOM as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
Claude Codedeclared by this listing
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Put your own value in place of <root> before you run this. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user dragoscv-axiom-mcp '--' npx -y @codai/axiom-mcp mcp --root '<root>'Reconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
Nothing is required, but it can take AXIOM_SIGNING_KEY (Optional Ed25519 private key (base64 PKCS#8 or raw seed) used by axiom_sign; never stored under a root.) if you want to set it.
What it writes
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
What you approve
19 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Validate a Plan against PlanSchema and, when all sources are inline, compute its planDigest. Read-only; touches no files.
No parameter schema on file.
Compile a Plan into a content-addressed ManifestBundle (sorted artifacts, sha256 digests, in-toto planDigest). store: cas writes blobs under <root>/.axiom/cas instead of inlining them. When a root is given the bundle is stored under <root>/.axiom/manifests/<hex>.json so later tools can reference…
No parameter schema on file.
Structural and content-address verification: schema, recomputed manifestDigest, every inline blob hashes to its key, attestation subject matches. When a root with .axiom/trust/keys.json is available, detached DSSE signatures are verified and the trusted keyids are reported under signatures.…
No parameter schema on file.
Evaluate the profile's predicates (plus the manifest's own checks) against the bundle. Repo facts are read from the root when one is available and the profile allows it. Verdict error means a provider could not run — never a silent pass.
No parameter schema on file.
Same evaluation as axiom_check, but returned immediately as a task so long guard.external suites (up to 15 min per guard) outlive the client's per-call timeout. Poll axiom_task_get with the returned taskId every pollIntervalMs until status is completed|failed|cancelled; the CheckReport is in…
No parameter schema on file.
Status of a task created by axiom_check_start. While working only the descriptor is returned; once terminal, result (completed) or error (failed|cancelled) is attached. Unknown or expired taskId → ERR_TASK_NOT_FOUND.
No parameter schema on file.
Abort a working task: every running guard process tree is killed and the task ends cancelled with error ERR_TASK_CANCELLED. Idempotent — a terminal task is returned unchanged.
No parameter schema on file.
Start assembling a Plan whose JSON would exceed the 4 MiB per-call cap: send the header here (everything except artifacts), append artifacts in chunks with axiom_plan_add, then axiom_plan_seal compiles the whole. The sealed bundle's manifestDigest is identical to a one-shot axiom_plan_compile of…
No parameter schema on file.
Add a chunk of Plan artifacts (each ≤ 4 MiB call, inline content ≤ 256 KiB per artifact as in a Plan) to an open session. Paths must be unique across every chunk (ERR_INVALID_PLAN otherwise); a sealed or over-budget session is ERR_PLAN_SESSION_STATE.
No parameter schema on file.
Assemble the session's header + every added artifact into one Plan and compile it exactly like axiom_plan_compile (same options, same digest). The session is consumed whether or not compilation succeeds. Bundles whose inline blobs exceed 4 MiB need store: cas (and therefore a root).
No parameter schema on file.
Stage the bundle under <root>/.axiom/staging, run pre-apply checks and produce a unified diff against the current tree. Nothing outside .axiom/ is touched. Echo the returned manifestDigest as confirmDigest to axiom_apply.
No parameter schema on file.
Transactionally write the bundle into the root: pre-image verification, staging, journal, atomic renames, scoped rollback on failure. Requires confirmDigest === bundle.manifestDigest (echo the digest you saw in dry-run). Idempotent: re-applying an applied digest is a no-op.
No parameter schema on file.
Replay the journal of a committed/committing manifest in reverse: restore backups, remove created files, drop the applied marker.
No parameter schema on file.
Compare two manifests by artifact path and digest. Each side is a ManifestBundle or a sha256:<hex> reference to a bundle stored under an allowlisted root.
No parameter schema on file.
Parse .axm v2 text into a Plan with 1-based {line, column} diagnostics; plan is present only when error-free. Read-only.
No parameter schema on file.
The frozen set of roots this server may read and write, as given by --root at startup.
No parameter schema on file.
Read-only snapshot of a root's write state, same as axiom status --json: who holds .axiom/lock (pid, host, digest, since, stale), the wait queue, registered in-flight intents (paths), interrupted journals that need axiom_rollback, the last applied chain entry and whether the journal hash chain…
No parameter schema on file.
Deterministic, content-addressed inventory of a root: every regular file (and symlink) as { path, bytes, sha256, mode, kind }, sorted by code point, with snapshotDigest = sha256(JCS(body)). No timestamps, no absolute paths — the same tree gives the same digest on every machine. Honours the root…
No parameter schema on file.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install AXIOM or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATrust score 94 / 100
Release 2.6.1 · scanned 14 Oct 2026
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score94
Findings10 threat · 0 capability
Capabilityhigh
Coveragesource
Verificationsource
Surface digeste2fc8cc8664ab05eb66209a820070fe0bb6653c3aff269c788bf6c1c3baf0433
Score vectorthreat 100 → client 94
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | source | publisher verification (provenance) — cryptographic build provenance ties the artifact to its source | −0 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (19 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| axiom_plan_validate | no tags | |
| axiom_plan_compile | state-write | state-write: declares readOnlyHint: false |
| axiom_manifest_verify | no tags | |
| axiom_check | no tags | |
| axiom_check_start | no tags | |
| axiom_task_get | no tags | |
| axiom_task_cancel | state-write | state-write: declares readOnlyHint: false |
| axiom_plan_begin | state-write | state-write: declares readOnlyHint: false |
| axiom_plan_add | state-write | state-write: declares readOnlyHint: false |
| axiom_plan_seal | state-write | state-write: declares readOnlyHint: false |
| axiom_apply_dry_run | no tags | |
| axiom_apply | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| axiom_write | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| axiom_rollback | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| axiom_manifest_diff | no tags | |
| axiom_axm_parse | no tags | |
| axiom_roots_list | no tags | |
| axiom_status | no tags | |
| axiom_repo_snapshot | no tags |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digeste2fc8cc8664ab05e
Scanned2026-10-14
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 2.6.1. Pin a release below and the command asks for exactly that one.
2.6.1
14 Oct 2026
latestNo release note was published with this version.
2.6.0
14 Oct 2026
No release note was published with this version.
2.5.0
14 Oct 2026
No release note was published with this version.
2.4.0
27 Sep 2026
No release note was published with this version.
2.3.0
23 Sep 2026
No release note was published with this version.
2.2.1
undated
- The version this listing was on when mcprush began following its releases. The registry was not asked when it shipped, so this row carries no date.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish AXIOM can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.