MCP Server for Z.AI - A Model Context Protocol server that provides AI capabilities
Nothing to run from here
Session Relay 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.
93/100
trust score
49
downloads a day
local
runs on your machine
2k
downloads in 30d
What it does
Share coding-agent sessions with your team: /share in one agent, /pickup in another.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user ryrahul-session-relay-mcp '--' npx -y session-relay mcp
# 2 — ask your agent for something these tools do
# list_sessions, search_sessions, get_session and 12 more
Session Relay costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
RYAll 1 server
Ryrahul1 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 Session Relay 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. 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 ryrahul-session-relay-mcp '--' npx -y session-relay mcpReconnect, 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 RELAY_SERVER (URL of your team's Session Relay server. Optional if you ran relay login.), RELAY_TOKEN (Personal access token (srk_...) from Settings → Tokens. Optional if you ran relay login.) if you want to set them.
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
15 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.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
List sessions teammates have shared. Defaults to the current repo; pass all_repos to see everything. for_me lists handoffs assigned to you (from any repo).
| Parameter | Type | Required | Notes |
|---|---|---|---|
| author | string | no | Name, email or "me" |
| tool | no | Only sessions from this coding agent | |
| branch | string | no | Only sessions shared from this git branch |
| file | string | no | Only sessions that touched this file (part of the path is fine) |
| for_me | boolean | no | Only handoffs assigned to you |
| all_repos | boolean | no | List sessions from every repo, not just the current one |
| limit | integer | no | How many to return, newest first (default 10) |
Full-text search over shared sessions (titles, summaries, messages). The query can include "file:<path>" and "branch:<name>" filters.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| query | string | yes | Words to look for, e.g. "invoice rounding" or "file:billing.ts retry" |
| this_repo_only | boolean | no | Limit to the current repo (default false) |
| author | string | no | Only sessions shared by this person: name, email or "me" |
Load a shared session as a handoff: summary, decisions, next steps, files and conversation. mode "tail" (default) = summary + recent messages; "summary" = summary only; "full" = full transcript, paged with offset.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | Session id from list_sessions or search_sessions |
| mode | string | no | How much to load (default "tail") |
| tail | integer | no | With mode "tail": only this many of the most recent messages (default: all that were shared) |
| offset | integer | no | With mode "full": message to start from, for paging |
Share the session you are in right now with the team. Write the summary yourself from the conversation: it is what teammates read first. Never include secrets.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| goal | string | yes | What this session is trying to achieve |
| decisions | array | no | Key decisions and why |
| done | array | no | What was finished in this session |
| next | array | no | Concrete next steps for whoever continues |
| open_questions | array | no | Unresolved questions the next person should know about |
| title | string | no | Short title; defaults to the first prompt (or the goal, when the user's prompts are left out) |
| include | string | no | What to upload besides your summary: nothing, the last 30 messages, or the whole transcript. Default "summary+tail" |
| people | array | no | Names or emails who can see it; omit for the whole team |
| continues | string | no | Id of the shared session this work continues. Defaults to the one loaded with get_session in this repo; "none" to unlink. |
| sections | array | no | Extra sections, e.g. the ones share_template asks for |
| include_diff | boolean | no | Attach the uncommitted changes in this repo (redacted), so the next person gets the exact code state |
| ignore_template | boolean | no | Share even if the repo's template has required sections you left empty |
| for | string | no | Name or email of the teammate this handoff is for |
| pr_url | string | no | Pull request URL. Defaults to the current branch's open PR (via gh); "none" to leave out |
| hide_prompts | boolean | no | Leave the user's typed prompts out (a marker takes their place); your replies and tool steps are kept. Pass true when the user asks to share without their messages. Defaults to the user's setting (rel |
The uncommitted changes a teammate attached to a shared session, as a unified diff. Use it to see exact code state; to apply it, write it to a file and run git apply.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | Session id |
| max_chars | integer | no | Cut the diff at this many characters (default 100000) |
The handoff template for the current repo: extra sections (some required) and instructions to follow when sharing. Call it before share_current_session.
No parameter schema on file.
What the team did recently: new shares by repo, handoffs for you, who picked up what, and new comments.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| days | integer | no | Look back this many days (default 1) |
Leave a short note on a shared session for the team: a question, a finding, or that you picked it up. Never include secrets.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | Session id |
| body | string | yes | The note, in plain text or Markdown |
Change a share you made (or any share, if you're an admin): title, summary fields, who it's for, the PR link or visibility. Summary lists you pass replace the old ones; use add_next to append next steps.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | yes | Id of the session to change |
| title | string | no | New title |
| goal | string | no | New goal |
| decisions | array | no | Replaces the decisions |
| done | array | no | Replaces what was done |
| next | array | no | Replaces the next steps |
| add_next | array | no | Next steps to append to the existing ones |
| open_questions | array | no | Replaces the open questions |
| sections | array | no | Replaces the extra sections |
| for | string | no | Teammate name or email, or "none" |
| pr_url | string | no | Pull request URL, or "none" |
| people | array | no | Make it visible only to these people; an empty list makes it visible to the whole team |
| publish | boolean | no | Publish a draft (saved automatically at session end) to the team, or to people |
Start working live in this repo: teammates see what you're working on and which files you touch, and their messages reach you between steps. Lists who else is live here.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| title | string | no | What this session is working on, in a few words |
| autonomy | boolean | no | Let this agent act on teammates' messages without asking the user first. Only true if the user explicitly asked for it. |
Who is working live right now: repo, branch, what they're doing, files they touched recently, and whether this session is live.
No parameter schema on file.
Send a short message to a teammate who is working live: it reaches their agent between steps. Use it to tell them about a change that affects them, ask a question, or answer theirs. Never include secrets.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| to | string | no | Teammate name or email, "everyone" (all live in this repo), or a live session id. Optional when replying. |
| message | string | yes | The message, in plain words |
| kind | string | no | "question" asks them to answer; "answer" replies to one (with reply_to) |
| reply_to | integer | no | Number of the message you're answering |
| wait_seconds | integer | no | For questions: wait up to this long for the answer and return it (default 0, don't wait) |
Take new messages from teammates sent to this live session (they are marked as read).
No parameter schema on file.
Change what teammates see about this live session (title, current status), or whether this agent may act on their messages without asking.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| title | string | no | What this session is working on |
| status | string | no | What you're doing right now, e.g. "running the migration tests" |
| autonomy | boolean | no | true only if the user explicitly asked to let you act on teammates' messages without asking |
End this session's live session: teammates stop seeing it and messages stop arriving.
No parameter schema on file.
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under AGPL-3.0
- 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 Session Relay 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 93 / 100
Release 0.7.0 · scanned 28 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 score93
Findings1 threat · 3 capability
Capabilityhigh
Coveragesource
Verificationrepo
Surface digest4868b1f5937a9f2fdc0428a452163194f201d020fd420c214875c73e6d196315
Score vectorthreat 100 → client 93
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| 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 (15 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 |
|---|---|---|
| list_sessions | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| search_sessions | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| get_session | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server), handler fetches a URL taken from its input (client.get(id) at dist/relay.js:3297 |
| share_current_session | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| get_session_diff | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| share_template | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| team_activity | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| comment_on_session | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| update_session | external-sinkuntrusted-input | external-sink: declares readOnlyHint: false, destructiveHint: true, openWorldHint: true; untrusted-input: handler fetches a URL taken from its input (client.get(args2.id) at dist/relay.js:3496 |
| start_live | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| live_status | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| send_live_message | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| check_live_inbox | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| update_live | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
| stop_live | external-sink | external-sink: declares readOnlyHint: false, openWorldHint: true |
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.
flow-1
Untrusted input can drive an external action even though no sensitive source is exposed.
Supply chain and provenance
Integrityfirst-seen
Surface digest4868b1f5937a9f2f
Scanned2026-10-28
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 0.7.0. Pin a release below and the command asks for exactly that one.
0.7.0
undated
latestNo release note was published with this version.
—
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 Session Relay 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.