Skip to content
verify mcp Beta VerifyMCP is currently in beta. If you notice any issues, get in touch and we’ll put it right.

AgentGuard

NPM · @AGENTGUARD-RUN/MCP

Local spend caps, kill-switch blocking and Ed25519-signed receipts for AI agent tool calls.

Available components

0 this week 68 Trust /100

Last scanned (15 hours ago)

Scores higher than 39% of npm servers

Category Security

Trust breakdown (7 categories)

How this component scores in each security and reliability category. Every signal is checked automatically from public evidence about the published package, including repeated runs of it in an isolated sandbox, and we only credit what we can confirm. How we score → Why this is hard to score →

Supply Chain Security81
  • No malware found by supply-chain analysis.Pass
  • CVE check failed: a known high-severity CVE affects zod 3.25.76, a direct dependency. A fixed version is available. View diagnostics → Fail
  • No install/post-install scripts declared.Pass
  • 31 of 97 dependencies flagged as unhealthy. View diagnostics → Partial
Provenance & Transparency6
  • Repository check failed: the declared repository URL redirects; it must resolve directly. See how to fix → View diagnostics → Fail
  • Provenance check failed: no build-provenance attestation is published. See how to fix → View diagnostics → Fail
  • Licence could not be checked: the licence is shipped as a file inside the package, not declared as a standard (SPDX) licence identifier, so we cannot tell whether it is OSI-approved.Unverified
  • Actively maintained (last published 60 days ago).Pass
  • Disclosure check failed: no security disclosure policy was found in the source repository. See how to fix → Fail
Schema Quality & AI Usability63
  • AI-judged instruction clarity (good).Pass
  • Context-footprint check failed: tool/resource definitions use about 1021 tokens (~170/item across 6 items; 6 tools + 0 resources), over budget; trim descriptions and params. See how to fix → Fail
  • Usage-examples check failed: none of the tools include examples. See how to fix → Fail
Stability & Change Management97
  • Stability observed for 29 of 30 days with no destabilising changes; credit accrues until the full window elapses.Partial
Tool Coverage94
  • 100% of tools have a non-trivial description (not blank, and not just the tool's name).Pass
  • 83% of tool parameters carry a description.Partial
Tool Safety100
  • No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.Pass
  • We read all 6 captured tool definition(s), and no name or description among them implies an irreversible operation.Pass
  • An AI judge read all 6 captured unit(s) of tool text and found none that tries to manipulate the model reading it.Pass
Capabilities100
  • Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.Pass
What we tested

Checks we ran (15)

  • Malware scan
  • Known CVEs
  • Install scripts
  • Dependency health
  • Source repository
  • Provenance
  • Maintenance
  • Security disclosure
  • Schema quality
  • Stability
  • Tool coverage
  • Injection markers
  • Destructive annotations
  • Judged manipulation
  • MCP protocol

Checks we could not complete (1)

These are gaps on our side, not findings about this package. We only credit what we can confirm, so a category made up entirely of them scores 0.

  • License in Provenance & TransparencyLicence could not be checked: the licence is shipped as a file inside the package, not declared as a standard (SPDX) licence identifier, so we cannot tell whether it is OSI-approved.

Does not apply to this package (2)

  • Endpoint SecurityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.
  • Transport & ReachabilityScored on hosted endpoints only. A package has no live endpoint of its own for us to probe.

The score does not measure everything. What the score does not measure →

Changelog

Every change we have recorded for this component, newest first. Security-relevant changes are always shown. ▲ marks a change for the better, ▼ a change for the worse; unmarked changes are neutral.

  • 21 Nov 26 0
    • Stability: pass → 0.97 functional
  • 15 Nov 26 0
    • Stability: 0.97 → pass security
  • 14 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 93 to 97. That category is still filling its 30-day observation window: 28 days of observed history at the previous scan, 29 at this one. The score rises as the window fills, whether or not the server changes.

  • 11 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 83 to 87. That category is still filling its 30-day observation window: 25 days of observed history at the previous scan, 26 at this one. The score rises as the window fills, whether or not the server changes.

  • 9 Nov 26 +1
    • Source repository: unverified → fail ▼ security
  • 8 Nov 26 0
    • Provenance: fail → unverified ▼ security
  • 7 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 70 to 73. That category is still filling its 30-day observation window: 21 days of observed history at the previous scan, 22 at this one. The score rises as the window fills, whether or not the server changes.

  • 5 Nov 26 +1

    No change was recorded against any check on this day. Stability & Change Management went from 63 to 67. That category is still filling its 30-day observation window: 19 days of observed history at the previous scan, 20 at this one. The score rises as the window fills, whether or not the server changes.

Diagnostics

Diagnostic detail from the automated scan of this channel: what the scanner observed at each step, so you can see exactly where a check passed or failed. It is informational only and never changes the trust score.

Captured 21 Nov 2026 · Analysed npm/@agentguard-run/mcp@0.3.1

Provenance No attestation

The registry publishes no build provenance for this version, so there is nothing to verify.

Result No attestation
Ecosystem npm

Background: How many MCP packages publish verified provenance →

Vulnerabilities 1 finding
ID CVE Severity Vector Fix available
GHSA-mrhx-c7xc-m3w5 CVE-2023-54404 high CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H yes

Background: What a vulnerability scan can and cannot prove →

Dependencies 97 packages
Packages resolved 97
Stale 31
Tree resolution Complete

Background: SBOMs and build attestations, explained →

MCP tools · 6 exposed · ~1,021 tokens

The tools this component advertises to a client, with an estimated token cost for each. Expand a tool to see its parameters and schema. The per-tool counts are indicative and are not scored directly; the schema's total context footprint is one signal in Schema Quality & AI Usability. A tool's description is untrusted text the model reads on every call, which is what makes this list a security surface and not just an inventory: how tool poisoning works →

Tool Tokens
export_receipts ~140

Export the tenant decision ledger, including earlier process runs, as a hash-chained array of signed receipts with a chain verification result. Memory mode exports this process only. Optional agent_id, task_id and workflow_id accept identifiers only; spend_decide binds them into the signed actor. Read-only and cost-configuration tools accept these fields without recording a spend decision.

NameTypeReqDescription
agent_idstring–Optional agent identifier; signed as actor.agentId on spend_decide
task_id––Optional task identifier; signed as actor.taskId on spend_decide
workflow_id––Optional workflow identifier; signed as actor.workflowId on spend_decide

No output schema declared.

No examples provided.

provenance_preview ~154

Preview the provenance AgentGuard would attest for a model and endpoint WITHOUT spending: model identity, weights origin country (China-origin families flagged), hosting jurisdiction, retention posture. Optional agent_id, task_id and workflow_id accept identifiers only; spend_decide binds them into the signed actor. Read-only and cost-configuration tools accept these fields without recording a spend decision.

NameTypeReqDescription
agent_idstring–Optional agent identifier; signed as actor.agentId on spend_decide
endpoint_urlstring––
modelstringyes–
task_id––Optional task identifier; signed as actor.taskId on spend_decide
workflow_id––Optional workflow identifier; signed as actor.workflowId on spend_decide

No output schema declared.

No examples provided.

set_model_cost ~176

Register a cost for a model with no built-in pricing (required for self-hosted models: AgentGuard fails closed on unknown costs). Cents per 1,000 tokens. Optional agent_id, task_id and workflow_id accept identifiers only; spend_decide binds them into the signed actor. Read-only and cost-configuration tools accept these fields without recording a spend decision.

NameTypeReqDescription
agent_idstring–Optional agent identifier; signed as actor.agentId on spend_decide
input_cents_per_ktoknumberyes–
modelstringyes–
output_cents_per_ktoknumberyes–
task_id––Optional task identifier; signed as actor.taskId on spend_decide
workflow_id––Optional workflow identifier; signed as actor.workflowId on spend_decide

No output schema declared.

No examples provided.

spend_decide ~251

Gate a model call BEFORE it runs. Pass metadata only (model, token counts, endpoint URL), never prompt content. Returns allow/block plus an Ed25519-signed, content-free receipt with model and hosting provenance (weights origin, jurisdiction, retention). Unknown self-hosted models need set_model_cost first (fail-closed). Optional agent_id, task_id and workflow_id accept identifiers only; spend_decide binds them into the signed actor. Read-only and cost-configuration tools accept these fields without recording a spend decision.

NameTypeReqDescription
agent_idstring–Optional agent identifier; signed as actor.agentId on spend_decide
endpoint_urlstring–Endpoint URL. Unrecognized endpoints route as self-hosted (zero retention, your jurisdiction).
input_tokensintegeryesProjected input tokens
modelstringyesModel name, e.g. llama-3.1-8b-instruct or gpt-5-mini
output_tokensintegeryesProjected output tokens
task_id––Optional task identifier; signed as actor.taskId on spend_decide
workflow_id––Optional workflow identifier; signed as actor.workflowId on spend_decide

No output schema declared.

No examples provided.

spend_status ~129

Show the active policy: caps, tenant, enforcement mode, decisions so far this session, and the signer identity. Optional agent_id, task_id and workflow_id accept identifiers only; spend_decide binds them into the signed actor. Read-only and cost-configuration tools accept these fields without recording a spend decision.

NameTypeReqDescription
agent_idstring–Optional agent identifier; signed as actor.agentId on spend_decide
task_id––Optional task identifier; signed as actor.taskId on spend_decide
workflow_id––Optional workflow identifier; signed as actor.workflowId on spend_decide

No output schema declared.

No examples provided.

verify_receipt ~171

Independently verify an AgentGuard signed receipt. Returns valid true/false. Tampering with any signed field makes verification fail. Optional agent_id, task_id and workflow_id accept identifiers only; spend_decide binds them into the signed actor. Read-only and cost-configuration tools accept these fields without recording a spend decision.

NameTypeReqDescription
agent_idstring–Optional agent identifier; signed as actor.agentId on spend_decide
public_key_hexstring–Signer public key (64 hex characters). Defaults to this server signer.
receipt_jsonstringyesThe signed receipt entry as a JSON string
task_id––Optional task identifier; signed as actor.taskId on spend_decide
workflow_id––Optional workflow identifier; signed as actor.workflowId on spend_decide

No output schema declared.

No examples provided.

Install

How do I install the AgentGuard MCP server?

AgentGuard runs locally as an npm package, launched with npx -y @agentguard-run/mcp. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation.

npm · @agentguard-run/mcp

# add to Claude Code
claude mcp add merchantguard-agentguard -- npx -y @agentguard-run/mcp
Common questions

What is the AgentGuard MCP server?

AgentGuard is an MCP server listed in the public MCP registry as io.github.MerchantGuard/agentguard. Local spend caps, kill-switch blocking and Ed25519-signed receipts for AI agent tool calls. This page covers its npm package (@agentguard-run/mcp).

Is the AgentGuard MCP server safe to use?

AgentGuard scores 68 out of 100 on VerifyMCP. We recorded 1 known advisory against it as of 21 November 2026. It declares no install or post-install scripts. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm.

What tools does the AgentGuard MCP server expose?

AgentGuard exposes 6 tools: spend_decide, verify_receipt, export_receipts, set_model_cost, provenance_preview, spend_status. Their descriptions and schemas cost roughly 1,021 tokens of context every time the server is loaded.

Is the AgentGuard MCP server still maintained?

AgentGuard is still listed as active in the MCP registry. We last reached this channel on 21 November 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

What did VerifyMCP test on the AgentGuard MCP server?

VerifyMCP ran 15 checks on AgentGuard, covering Malware scan, Known CVEs, Install scripts, Dependency health and 11 more. One item could not be completed, a gap on our side rather than a finding about AgentGuard: License. Endpoint Security and Transport & Reachability do not apply to an npm package, so they were not tested. The score does not measure everything, and the scoring documentation lists what it leaves out.