MCP.Market

Policyvault MCP server

by zapsoblige-hash·io.github.zapsoblige-hash/policyvault·v1.6.2·1 stars

Non-custodial Kaspa policy enforcement for AI agents; signers retain custody.

B80/100grade B
What users say
No reviews yet
Be the first
Safety scan
B80/100

full report

Adoption
Growing

1 stars50 downloads/wk

Reviews

Write one

Nobody has reviewed Policyvault yet.

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

Policyvault tools (19, 3 write)

write = sends, deletes, buys or posts

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

  • policyvault_audit_feed

    Read the global activity feed, server-scoped to vaults this identity's creating wallet participates in. Requires scope read:audit. Read-only.

  • policyvault_capabilities

    Read the server's public capability/version discovery document (API version, network, supported covenant versions, v0.4 actions, scope enum, schema versions, limits, feature flags). Read-only; requires no scope.

  • policyvault_create_org_root_requestwrite action

    BUILD a durable v0.7 root-authorized request: a root-only owner action (authorize/rotate/freeze/unfreeze/ownerRecover/succession) or the same action carrying AT MOST ONE rooted-vault operation (the covenant itself refuses a second rooted vault as a foreign covenant rider — ONE_VAULT_OPERATION_PER_ROOT_TRANSITION otherwise). This NEVER signs and NEVER broadcasts: it produces per-owner-slot signer-r

  • policyvault_create_requestwrite action

    BUILD a durable v0.4 wallet request (exact unsigned transaction + canonical review) for an existing vault. This NEVER signs, NEVER broadcasts, and cannot move funds by itself: signatures come from external signer custody and submission is a separate, separately-scoped step outside this tool set. Amounts are integer sompi as decimal strings — never floats. Server-side governance (proposalId for aut

  • policyvault_create_v7_requestwrite action

    BUILD a durable v0.7 rooted-vault request: a delegate spend (tokenAgentSpend on a rooted token vault, or agentSpend on a rooted KAS safe-payment vault — an authorized agent's own key; above the agent's approvalThreshold the vault-level approvers co-sign through the server's approvals route) or a token deposit — NEITHER touches the organizational root at all (no root input; the covenant enforces th

  • policyvault_governance_proposal

    Read one governance proposal (classification, quorum, collected approvals, consumption state) by proposalId. Requires scope read:governance. Read-only.

  • policyvault_governance_proposals

    List governance proposals (authority-expansion ceremonies with owner-signature approvals) visible to this identity, optionally narrowed to one vault. Requires scope read:governance. Read-only — proposal creation/approval/cancellation are human ceremonies outside this tool set.

  • policyvault_list_requests

    List durable v0.4 wallet requests visible to this identity (optionally narrowed to one vault, optionally only open/actionable states). Requires scope read:requests. Read-only.

  • policyvault_list_vaults

    List the vaults this machine identity's creating wallet participates in (tenancy is enforced server-side; foreign vaults are invisible). Requires scope read:vaults. Read-only.

  • policyvault_network_status

    Read the configured Kaspa node's status (network id, sync state, UTXO index, virtual DAA score). Requires scope read:network. Read-only.

  • policyvault_org_root

    Read one v0.7 organizational root's full record (owner slots, thresholds M/K/R, freeze flag, generation/nonce, pending request, live outpoint) plus authorityModel and a human explanation, by rootCovenantId. A root not yet chain-proven answers 404. Requires scope read:org-roots. Read-only.

  • policyvault_org_root_requests

    List every durable request against one organizational root (root genesis, rooted-vault genesis, or a root-authorized owner action), or read one by requestId — per-slot signature status, requiredApprovals/signaturesPresent, the outcome ladder, and warnings. Requires scope read:org-roots. Read-only.

  • policyvault_org_roots

    List v0.7 ON-CHAIN ORGANIZATIONAL ROOTS visible to this identity (consensus-enforced M-of-N owner authority — distinct from a HOSTED organization, which grants NO on-chain authority). Every summary carries authorityModel. Requires scope read:org-roots. Read-only.

  • policyvault_reject_request

    Mark one open wallet request REJECTED (housekeeping for a stale or superseded build — e.g. after a policy change made it unsubmittable). Affects only the durable workflow record; nothing on-chain. Requires scope request:reject.

  • policyvault_request_status

    Read one durable wallet request (state machine BUILT/AWAITING_APPROVALS/FINALIZED/SUBMITTED/CONFIRMED/REJECTED/FAILED, canonical review, manifest hash, txid once chain-proven) by requestId. Requires scope read:requests. Read-only.

  • policyvault_risk_evaluation

    Read one durable risk-evaluation record (decision ALLOW/REVIEW/DENY, codes, adapter evidence, release state) by evaluationId. Requires scope read:risk. Read-only — releasing a REVIEW hold is a human reviewer action outside this tool set.

  • policyvault_simulate_request

    DRY-RUN a v0.4 wallet operation through the real pipeline (governance classification, risk composition, authorization, exact build with real fee/mass, intent-manifest verification) WITHOUT persisting anything, consuming any gate, or broadcasting. Amounts are integer sompi as decimal strings (1 KAS = 100000000 sompi) — never floats. Requires scope request:build (ownerPause/ownerRecover additionally

  • policyvault_vault

    Read one vault's manifest detail (policy template, agent registry, live covenant state, operational status) by 64-hex vaultId. Requires scope read:vaults. Read-only.

  • policyvault_vault_audit

    Read the durable audit/activity events recorded for one vault (builds, approvals, submissions, governance and risk outcomes, reconciliation). Requires scope read:vaults. Read-only.

Public scan report

scanner v0.1.10 · 2026-10-03 · same rubric, same numbers if you re-run it

no findings
  • Code scan11 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 qualitystatic API keys via environment variables6/15
  • Maintenancelast push 4 days ago15/15
  • Maintainer identityregistry namespace matches repository owner6/10
Overall 80/100. Components that don't apply are left out of the denominator. Any critical finding is an F.RubricAppeal a findingJSON

How Policyvault compares

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

Safety score
80/100
higher than 60% of 48,255 servers
Tools
19
the typical server has 7
Tools that change things
16%
typical: 0%. Fewer write tools means less an agent can break
GitHub stars
1
more than 65% of 28,688 servers
Downloads a week
50
more than 42% of 12,191 servers with a package

What its tools let an agent do: 16 read · 3 change

Install Policyvault in Claude Code, Cursor or VS Code

Runs npx -y policyvault-mcp on your machine. Read the scan report first; the gateway never runs local packages.

claude mcp add policyvault -- npx -y policyvault-mcp
Add to Cursor

What the publisher says

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

PolicyVault

Non-custodial delegated-spending vaults on Kaspa L1 — for people and AI agents.

PolicyVault lets a vault owner hand a spending key to an agent — an employee, a service, a bot, or an AI agent — without handing over control of the funds. The spending policy is enforced by Kaspa L1 consensus through a covenant: even an agent who bypasses this entire application and talks directly to a Kaspa node cannot exceed the owner's policy.

The authority model, stated once and everywhere enforced:

**AI MAY REQUEST. POLICYVAULT DETERMINISTICALLY DECIDES.

THE COVENANT ENFORCES. SIGNERS RETAIN CUSTODY.**

Production status (honest labels)

Shortened. The full README is on GitHub.

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

Policyvault: common questions

Is Policyvault MCP server safe?
Mostly: it is graded B (80/100). Read the Policyvault safety report
How do I install Policyvault?
It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.
Does Policyvault need an API key?
Yes. The registry entry asks for POLICYVAULT_MCP_TOKEN.
Is Policyvault maintained?
The last commit was 36 days ago (2026-09-29). The latest release is v1.6.2.
How does Policyvault compare with other MCP servers?
safety score 80/100 (higher than 60% of 48,255 servers); tools 19 (the typical server has 7); tools that change things 16% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.