Gvnr

mightbesaad/gvnr

A hosted governance layer that sits outside your LLM calls and decides whether an agent is cleared to spend. You set USD spend envelopes per agent and window, then call `budget_clear` before each request and `reconcile` after with actual token counts. It also handles rate limits per model, idempotency keys to dedupe retries, and a human approval bridge when an agent hits a cap or needs sign-off. The governance operations quota is prepaid with USDC on Base at 1,000 ops per dollar, either through the web pay flow or x402. No proxy, no self-hosting. Useful when you want hard guardrails on agent spend without routing tokens through a gateway or waiting for the bill to know something went wrong.

tools: 10
auth: Secret needed
transport: HTTP
last updated: Jun 2, 2026

View on GitHub

Tools

10 tools

Public tool metadata for what this MCP can expose to an agent.

budget_clearPre-flight authorization for a planned LLM call. Estimates cost from model + estimated_tokens, deducts from the agent envelope, returns {approved:true, remaining_usd} or {approved:false, reason}. Call BEFORE the LLM request; if approved=false, skip the call. Pair with `reconci...3 params

Pre-flight authorization for a planned LLM call. Estimates cost from model + estimated_tokens, deducts from the agent envelope, returns {approved:true, remaining_usd} or {approved:false, reason}. Call BEFORE the LLM request; if approved=false, skip the call. Pair with `reconci...

Parameters* required

agent_idstring

The agent identifier

estimated_tokensinteger

Tokens to be billed — output tokens for chat models, input tokens for input-only models (embeddings)

modelstring

Model being called (e.g. claude-sonnet-4-6, gpt-4o, text-embedding-3-small)

set_envelopeCreate or update a per-agent USD spend envelope. Idempotent: re-calling preserves the running spent_usd and reset_at; only limit_usd and window are overwritten. `window:"daily"` resets at UTC midnight; `window:"session"` never resets (caller-managed). Must be set before `budge...3 params

Create or update a per-agent USD spend envelope. Idempotent: re-calling preserves the running spent_usd and reset_at; only limit_usd and window are overwritten. `window:"daily"` resets at UTC midnight; `window:"session"` never resets (caller-managed). Must be set before `budge...

Parameters* required

agent_idstring

The agent identifier

limit_usdnumber

Spend limit in USD

windowstring

Reset window: daily (UTC midnight) or session (never resets)one of daily · sessiondefault: daily

get_balanceRead-only snapshot of the account-level governance-operation quota. The quota is increased by pay-as-you-go USDC topups (any amount, credited at 1,000 ops per $1) and decreased by one per `budget_clear`. Your LLM token spend is billed by your provider, not here; per-agent spen...

Read-only snapshot of the account-level governance-operation quota. The quota is increased by pay-as-you-go USDC topups (any amount, credited at 1,000 ops per $1) and decreased by one per `budget_clear`. Your LLM token spend is billed by your provider, not here; per-agent spen...

No parameter schema in public metadata yet.

reconcilePost-call drift correction. After the LLM returns, call this with the actual input/output token counts from the provider response — applies the delta (actual minus estimated cost) to the agent envelope (the spend cap); the operation quota is unchanged. Pairs with `budget_clear...3 params

Post-call drift correction. After the LLM returns, call this with the actual input/output token counts from the provider response — applies the delta (actual minus estimated cost) to the agent envelope (the spend cap); the operation quota is unchanged. Pairs with `budget_clear...

Parameters* required

actual_input_tokensinteger

Actual input tokens reported by the LLM provider

actual_output_tokensinteger

Actual output tokens reported by the LLM provider

agent_idstring

The agent identifier

set_rate_envelopeCreate or update a rate-limit envelope scoped to the (agent_id, provider, model) triple. Each triple gets its own independent counter — different agents on the same model do not share quota. Fixed 60-second window (not sliding). Idempotent: re-calling updates requests_per_minu...4 params

Create or update a rate-limit envelope scoped to the (agent_id, provider, model) triple. Each triple gets its own independent counter — different agents on the same model do not share quota. Fixed 60-second window (not sliding). Idempotent: re-calling updates requests_per_minu...

Parameters* required

agent_idstring

The agent identifier

modelstring

Model identifier, e.g. claude-sonnet-4-6, gpt-4o

providerstring

Provider name, e.g. anthropic, openai, bedrock

requests_per_minuteinteger

Allowed requests per 60-second window

rate_checkCheck the rate-limit envelope for an (agent_id, provider, model) triple before making an LLM call. Returns {allowed:true, requests_remaining_this_minute} if under cap (and increments the counter), or {allowed:false, reason:"rate_exceeded", retry_after_ms} if over. Pairs with `...3 params

Check the rate-limit envelope for an (agent_id, provider, model) triple before making an LLM call. Returns {allowed:true, requests_remaining_this_minute} if under cap (and increments the counter), or {allowed:false, reason:"rate_exceeded", retry_after_ms} if over. Pairs with `...

Parameters* required

agent_idstring

The agent identifier

modelstring

Model identifier, e.g. claude-sonnet-4-6

providerstring

Provider name, e.g. anthropic, openai

idempotency_checkDedupe retries on a caller-supplied key. Returns {is_first_call:true} the first time a key is seen within TTL; returns {is_first_call:false} on every subsequent call. Use as a guard around non-idempotent operations — common pairings: gate `reconcile` to prevent double-correcti...2 params

Dedupe retries on a caller-supplied key. Returns {is_first_call:true} the first time a key is seen within TTL; returns {is_first_call:false} on every subsequent call. Use as a guard around non-idempotent operations — common pairings: gate `reconcile` to prevent double-correcti...

Parameters* required

keystring

Idempotency key — unique per logical operation

ttl_secondsinteger

Time-to-live in seconds (default 3600, max 30 days)

request_approvalRequest human approval for an agent action. Returns immediately with {approval_id, approval_url, expires_at}; the human approves or denies via the URL (mobile-first web page), and the agent polls `check_approval` with approval_id to learn the decision. Notifies the account hol...4 params

Request human approval for an agent action. Returns immediately with {approval_id, approval_url, expires_at}; the human approves or denies via the URL (mobile-first web page), and the agent polls `check_approval` with approval_id to learn the decision. Notifies the account hol...

Parameters* required

action_summarystring

Human-readable description of the action awaiting approval (≤280 chars)

agent_idstring

Agent identifier requesting approval

channelsarray

Notification channels (default ["email"]; telegram/sms accepted for forward-compat but not yet delivered)

ttl_secondsinteger

Time the approver has to decide (default 600s, max 7 days)

check_approvalPoll the status of a pending approval created by `request_approval`. Returns {decision: "pending"} while the human has not decided, {decision: "approved"} or {decision: "denied"} once they have, or {decision: "timeout"} if expires_at passed without a decision. Read-only and id...1 params

Poll the status of a pending approval created by `request_approval`. Returns {decision: "pending"} while the human has not decided, {decision: "approved"} or {decision: "denied"} once they have, or {decision: "timeout"} if expires_at passed without a decision. Read-only and id...

Parameters* required

approval_idstring

approval_id returned by request_approval

view_dashboardOpen an interactive, read-only spend dashboard for this account: an operation-quota gauge plus a live bar per agent showing spent vs. envelope limit, remaining USD, and daily-reset countdown, with a recent-clearance activity feed. Use when a human asks how their budget or an a...

Open an interactive, read-only spend dashboard for this account: an operation-quota gauge plus a live bar per agent showing spent vs. envelope limit, remaining USD, and daily-reset countdown, with a recent-clearance activity feed. Use when a human asks how their budget or an a...

No parameter schema in public metadata yet.

Readme

Gvnr

Spend caps, rate coordination, idempotency, and a human-in-the-loop gate for AI agents — enforced before the call, not after the invoice. One MCP endpoint, settled via x402 (USDC on Base). No proxy, no self-hosting, no infrastructure to deploy.

Listed on the Official MCP Registry as dev.gvnr/gvnr.

The problem

View the full README on GitHub

Details

Repository: github.com/mightbesaad/gvnr
Website: gvnr.dev
Category: AI & LLM Tools
Transport: HTTP
Auth: Required
Registry: MCP registry, active

Related AI & LLM Tools MCP servers

View all

Minuetminuetai/minuet-mcpMCP server exposing Minuet's ERC-8004 agent relationship graph.0stars

Hive Doctrine Mcpmizukaizen/hive-doctrine-mcpAgent knowledge marketplace. 105 products, 6 MCP tools, x402 payments on Base network.0stars

Myai Mcpmyaitoken/myai-mcpDecentralized AI inference on Base: OpenAI-compatible chat via community GPUs, paid in MYAI0stars

Agency Mcp Servernpupko/agency-mcp-serverOn-demand access to 150+ specialist AI agent templates — search, browse, and spawn agents0stars

Local Model Suitability MCPojaskord/local-model-suitability-mcpAI model router. Checks whether to use local Ollama or cloud inference before each API call.0stars

AgentNetoxgeneral/agentnetAgent-to-agent referral network. Discover, recommend, and refer users between AI agents via MCP.0stars