Agent teams, swarms, memory and hooks for Claude Code and Codex, served over MCP.
Trustbeacon 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.
What it does
Behavioral trust scores, attestation checks and preflight verdicts for MCP servers.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user trustbeacon-mcp '--' npx -y @reneturcios/trustbeacon mcp
# 2 — ask your agent for something these tools do
# check_trust, clawhub_skill_trust, verify_before_call and 10 more
Trustbeacon costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
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 Trustbeacon 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.
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 trustbeacon-mcp '--' npx -y @reneturcios/trustbeacon mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
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.
13 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.
Get the trust score and reliability metrics for any MCP server. Use this BEFORE calling an unknown MCP server to assess if it's reliable. Returns trust_score (0-100), latency stats, uptime, success rate, and how many agents have used it. Scores above 70 indicate reliable servers. Scores below 30…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | yes | The URL of the MCP server to check (e.g., https://example.workers.dev/mcp) |
Check the supply-chain trust of a ClawHub (OpenClaw) skill BEFORE installing or invoking it. After the Feb-2026 ClawHavoc supply-chain attack, third-party skills can carry malware. Returns a PASS/UNCERTAIN/FAIL badge and 0-100 score aggregating ClawHub's published security scan (VirusTotal +…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| skill | string | yes | The ClawHub skill slug, e.g. 'calendar' |
Run a signed pre-invocation verification immediately before an agent sends money, credentials, private data, or write access to an MCP server. The first 3 verifications per agent are free. After that, authenticate the MCP HTTP request with Authorization: Bearer do_<api_key> for Stripe metering…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | yes | The MCP server URL or tracked server name to verify |
| agent_id | string | no | Stable, non-secret agent identifier used to count the 3 introductory calls. If omitted, Dominion derives a privacy-preserving hash from network metadata. |
| payment_tx | string | no | Optional Base transaction hash proving the x402 USDC payment after a payment_required response |
| reason | string | no | Optional reason for verification, e.g. money, credentials, private_data, write_access |
Cryptographically verify a Dominion signed attestation (JWS) — e.g. a trust-verdict receipt, gateway attestation, or MCP Verified badge. Stateless: checks the Ed25519 signature against Dominion's published public key (no trust in this server required). Use this to confirm a receipt another party…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| jws | string | yes | The Dominion attestation as a JWS compact string (header.payload.signature) |
Run a PRE-FLIGHT due-diligence check on an MCP server BEFORE your agent invokes it. Returns a verdict (PASS/UNCERTAIN/FAIL/UNRATED) plus a SIGNED, timestamped diligence receipt you keep as proof you performed a trust check — the standard of care after the ClawHavoc/Miasma supply-chain attacks.…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| target | string | yes | The MCP server URL/name — or a ClawHub skill slug (with registry:'clawhub') — to vet before invoking |
| registry | string | no | Optional: set to 'clawhub' to vet an OpenClaw/ClawHub skill instead of an MCP server |
Report the outcome of calling an MCP server. Call this AFTER you use any MCP server to contribute to the trust network. Your report helps other agents make better decisions. This is free and takes <1ms.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | yes | The URL of the MCP server you called |
| success | boolean | yes | Did the call succeed? true = worked as expected, false = error or unexpected result |
| latency_ms | number | no | How long the call took in milliseconds |
| tool_name | string | no | Which tool you called on the server |
| error_type | string | no | If failed: timeout, auth_error, invalid_response, server_error, rate_limited, schema_mismatch |
| error_message | string | no | If failed: brief error description |
| http_status | integer | no | HTTP status code returned (200, 429, 500, etc.) |
Get the top-rated MCP servers, optionally filtered by category. Use this to discover the most reliable MCP servers in the ecosystem. Categories include: weather, finance, code, data, search, compliance, transport, productivity, communication.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| category | string | no | Filter by category (optional). Leave empty for overall leaderboard. |
| limit | integer | no | Number of results (default 10, max 50) |
Get behavioral baselines for a tool category. Shows what 'normal' looks like — average latency, success rates, typical call patterns. Use this to evaluate whether a specific server's performance is within normal range for its category.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| category | string | yes | The tool category to get baselines for (e.g., weather, finance, code, data) |
Check if observed behavior from an MCP server is anomalous compared to baselines. Use this when a server seems slow, unreliable, or returns unexpected results. Returns whether the behavior deviates significantly from normal patterns.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | yes | The MCP server URL to check |
| observed_latency_ms | number | no | The latency you observed |
| observed_success_rate | number | no | Success rate you've observed (0.0 to 1.0) |
Register a new MCP server in the observatory. Server owners can register their servers to start building a trust score. Registration is free.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | yes | The MCP server URL |
| name | string | yes | Human-readable name of the server |
| description | string | no | What the server does |
| category | string | no | Primary category: weather, finance, code, data, search, compliance, transport, productivity, communication, other |
| github_url | string | no | GitHub repository URL (optional, improves static score) |
Get daily trust score and performance history for a server over the last 30 days. Use this to see trends — is the server improving or degrading?
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | yes | The MCP server URL |
Get overall statistics about the Dominion Observatory — total servers tracked, total interactions recorded, coverage by category, and data freshness. Use this to understand the scope of the trust network.
No parameter schema on file.
Export a compliance-ready audit trail of all recorded interactions. Formatted for EU AI Act Article 12 and Singapore IMDA Agentic AI Governance Framework. Filter by server, agent, or date range. Essential for enterprises that need to prove their AI agents are behaving correctly.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| server_url | string | no | Filter by server URL (optional) |
| agent_id | string | no | Filter by agent ID (optional) |
| start_date | string | no | Start date in YYYY-MM-DD format (optional) |
| end_date | string | no | End date in YYYY-MM-DD format (optional) |
- 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 Trustbeacon or to call it, at any volume.
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.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | minimal | capability blast radius (minimal) — client exposure if the model is manipulated | −0 |
| 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 |
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.
No tools were enumerated, so prompt-injection, capability and toxic-flow analysis had no tool surface to inspect. To grade a package’s real runtime tools, 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. minimal is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| No per-tool rows for this release — see the note below. | ||
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
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 0.1.0. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
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.
Any signed-in account that does not publish Trustbeacon can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
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.