Asqav Mcp

Name: Asqav Mcp
Author: jagmarques

by jagmarquescommunityv0.3.7

An MCP server that enforces policy checks on agent actions, signs approvals and results, and produces verifiable audit trails. It also provides tools to manage agents, enforce tool-level policies, and scan MCP tool definitions for security issues.

From the registry: Stop a rogue AI agent and prove what it tried: signed receipts, tool policies, output verification.

GitHub Repository

Install & configure

Prerequisites

required
Install Python package
Install the MCP server from PyPI using pip.
```
$ pip install asqav-mcp
```
https://pypi.org/project/asqav-mcp/
required
Create an Asqav API key
Get an API key from Asqav and make it available to the server.
https://asqav.com

Install in your client

Paste into the AI you're chatting with — it will figure out the rest

Please install the `asqav-mcp` MCP server into my current AI client (that's you).

Required prerequisites (do these first if not already done):
- **Install Python package** — Install the MCP server from PyPI using pip. Run: `pip install asqav-mcp` (https://pypi.org/project/asqav-mcp/)
- **Create an Asqav API key** — Get an API key from Asqav and make it available to the server. (https://asqav.com)

Canonical MCP server config (stdio transport):
- command: `asqav-mcp`
- args: []
- required environment variables:
  - `ASQAV_API_KEY`: Asqav API key used to authenticate to the Asqav API. (example: `<your-asqav-api-key>`)
- optional environment variables:
  - `ASQAV_API_URL`: Base URL for the Asqav API. Defaults to the Asqav cloud API; set this when using a self-hosted deployment. (example: `<your-asqav-api-url>`)
  - `ASQAV_PROXY_TOOLS`: JSON string defining local enforcement policies for tools, including risk levels, approval requirements, rate limits, blocking, hiding, and optional tool endpoints. (example: `<your-json-tool-policy-map>`)

Note: The README states the server uses stdio transport and can be run directly as `asqav-mcp`. Docker and local development setup are also documented, but the simplest portable path is the PyPI install plus the `asqav-mcp` command.

Add this MCP server to my current client's config in the correct format for you. If you need secrets or credentials I haven't provided, ASK me — do not invent values or leave raw placeholders. After adding it, tell me how to verify the server is connected.

Note: The README states the server uses stdio transport and can be run directly as `asqav-mcp`. Docker and local development setup are also documented, but the simplest portable path is the PyPI install plus the `asqav-mcp` command.

Tools (15)

check_policy

Check whether an action is allowed by the organization's policies.

preflight_check

Combined agent status and policy check; returns CLEARED or NOT CLEARED with reasons.

sign_action

Create a signed, replayable audit record for an agent action.

verify_signature

Verify a created signature.

verify_output

Verify a signed output matches expected content by comparing the stored output hash against a fresh hash.

list_agents

List all registered AI agents.

get_agent

Get details for a specific agent.

gate_action

Pre-execution enforcement gate that checks policy, signs approval or denial, and returns the verdict.

complete_action

Report the outcome of a gate-approved action and bind it to the original approval as a bilateral receipt.

enforced_tool_call

Strong enforcement proxy that checks policy, rate limits, and approval requirements, then optionally forwards the call and signs request and response together.

create_tool_policy

Create or update a local enforcement policy for a tool.

list_tool_policies

List all active tool enforcement policies.