# AgentGo (npm · agentgo-mcp)

Run Codex and Claude Code on your computer from an MCP client, and track or continue their runs.

- Trust score: 65/100 (medium)
- Change this week: +26
- Registry status: active
- Liveness: live
- Owner verified: no
- Last scored: 2026-10-17

## Components

- npm · `agentgo-mcp`: 65/100 (this document), [markdown](https://verifymcp.io/servers/nishkalkashyap-agentgo/agentgo-mcp.md), [page](https://verifymcp.io/servers/nishkalkashyap-agentgo/agentgo-mcp)

## Channel facts

- Registry: `npm`
- Package: `agentgo-mcp`
- Version: `0.3.1`
- Transport: `stdio`

## Trust breakdown

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. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog)

Scored 2026-10-17.

- **Supply Chain Security**: 99/100
  - No malware found by supply-chain analysis.
  - No known CVEs affecting this package version or its production dependencies.
  - No install/post-install scripts declared.
  - 3 of 18 dependencies flagged as unhealthy.
- **Provenance & Transparency**: 45/100
  - Source repository is publicly reachable at the declared URL.
  - Provenance check failed: no build-provenance attestation is published.
  - Clear OSI-approved license (MIT).
  - Actively maintained (last published 10 days ago).
  - Disclosure check failed: no security disclosure policy was found in the source repository.
- **Schema Quality & AI Usability**: 59/100
  - AI-judged instruction clarity (fair).
  - Tool/resource definitions use about 892 tokens (~68/item across 13 items; 13 tools + 0 resources), lean.
  - Usage-examples check failed: none of the tools include examples.
- **Stability & Change Management**: 0/100
  - Stability not yet verified: not enough scan history yet (needs a 30-day window).
- **Tool Coverage**: 67/100
  - 100% of tools have a non-trivial description (not blank, and not just the tool's name).
  - 0% of tool parameters carry a description.
- **Tool Safety**: 100/100
  - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured.
  - All 1 tool(s) whose name or description implies an irreversible operation declare an MCP destructiveHint annotation.
  - An AI judge read all 13 captured unit(s) of tool text and found none that tries to manipulate the model reading it.
- **Capabilities**: 100/100
  - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28.

**Unverified: 1 category.** A category scored 0 because we could not verify it: a data source with nothing on this package, evidence we could not reach, or a check we could not run. We only credit what we can confirm.

## Install

### How do I install the AgentGo MCP server?

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

### Claude

```bash
claude mcp add nishkalkashyap-agentgo -- npx -y agentgo-mcp
```

### Cursor

```json
{
  "mcpServers": {
    "nishkalkashyap-agentgo": {
      "command": "npx",
      "args": [
        "-y",
        "agentgo-mcp"
      ]
    }
  }
}
```

### VS Code

```json
{
  "servers": {
    "nishkalkashyap-agentgo": {
      "command": "npx",
      "args": [
        "-y",
        "agentgo-mcp"
      ]
    }
  }
}
```

### Codex

```bash
codex mcp add nishkalkashyap-agentgo -- npx -y agentgo-mcp
```

### opencode

```json
{
  "$schema": "https://opencode.ai/config.json",
  "mcp": {
    "nishkalkashyap-agentgo": {
      "type": "local",
      "command": [
        "npx",
        "-y",
        "agentgo-mcp"
      ],
      "enabled": true
    }
  }
}
```

### OpenClaw

```bash
openclaw mcp add nishkalkashyap-agentgo --command npx --arg -y --arg agentgo-mcp
```

### Hermes

```yaml
mcp_servers:
  nishkalkashyap-agentgo:
    command: "npx"
    args: ["-y", "agentgo-mcp"]
```

### Netclaw

```json
{
  "McpServers": {
    "nishkalkashyap-agentgo": {
      "Transport": "stdio",
      "Command": "npx",
      "Arguments": [
        "-y",
        "agentgo-mcp"
      ]
    }
  }
}
```

### Vellum

```bash
assistant mcp add nishkalkashyap-agentgo -t stdio -c npx -a -y agentgo-mcp
```

### Other

```json
{
  "mcpServers": {
    "nishkalkashyap-agentgo": {
      "command": "npx",
      "args": [
        "-y",
        "agentgo-mcp"
      ]
    }
  }
}
```

## Changelog

Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims.

### 2026-10-14 (score 65, +26)

- [security improvement] Injection markers: unverified → pass
- [security] First check of Judged manipulation: pass
- [security] Stability: Stability not yet verified: not enough scan history yet (needs a 30-day window).
- [functional improvement] Tool coverage: unverified → 100
- [functional improvement] MCP protocol: unverified → pass
- [functional] First check of Tool coverage: 0
- [functional] First check of Schema quality: fail
- [functional] First check of Schema quality: fair
- [functional] First check of Destructive annotations: 100
- [functional] First check of Schema quality: pass

### 2026-10-13 (score 39, 0)

- [functional regression] Security disclosure: unverified → fail

### 2026-10-12 (score 39, +15)

- [security improvement] Malware scan: unverified → pass

### 2026-10-11 (score 24, −15)

- [security regression] Malware scan: pass → unverified
- [functional regression] Security disclosure: fail → unverified

### 2026-10-09 (score 39)

First indexed and scored.

## MCP tools (13)

### `getAgentCapabilities` (~44 tokens)

Show which coding agents (codex, claude) are installed on this computer, their versions, and how they approve their own actions.

Input parameters:

- `provider` (string)

### `getSupportedModels` (~68 tokens)

List the models you can pass to startAgentRun, with the effort and serviceTier values each accepts. Codex reports its list live; the Claude list is set by the owner and may include models their account cannot use.

Input parameters:

- `provider` (string, required)
- `refresh` (boolean)

### `listWorkspaces` (~57 tokens)

List the projects the owner has allowed agents to work in. Use an id as workspaceId in other tools. Only the owner can add projects, by running agentgo on their computer; if the list is empty, the result says how.

### `listDirectory` (~65 tokens)

List the files and folders at one path in a workspace. If nextCursor is set, pass it back as cursor for the next page.

Input parameters:

- `cursor` (string)
- `limit` (integer)
- `path` (string)
- `workspaceId` (string, required)

### `readFile` (~78 tokens)

Read lines from a UTF-8 text file in a workspace (files up to 1 MiB). Secrets such as .env files and private keys, and symlinks, are blocked.

Input parameters:

- `maxLines` (integer)
- `path` (string, required)
- `startLine` (integer)
- `workspaceId` (string, required)

### `globFiles` (~74 tokens)

Find workspace files whose paths match a glob such as src/**/*.ts. Skips files ignored by .gitignore, dependency and build folders such as node_modules and dist, and secrets.

Input parameters:

- `cursor` (string)
- `limit` (integer)
- `pattern` (string, required)
- `workspaceId` (string, required)

### `grepFiles` (~85 tokens)

Search workspace files with ripgrep and return matching lines with line numbers. Skips the same files as globFiles. If truncated is true, narrow the search.

Input parameters:

- `caseSensitive` (boolean)
- `fileGlob` (string)
- `limit` (integer)
- `literal` (boolean)
- `pattern` (string, required)
- `workspaceId` (string, required)

### `startAgentRun` (~164 tokens)

Start Codex or Claude Code on a task in a workspace. It runs in the background and approves its own actions. Several runs can work at once, even in the same workspace, so give parallel runs separate parts of the code. Returns a taskId to poll with getAgentRunStatus and getAgentRunOutput. If this call fails, retry with the same arguments and idempotencyKey; that never starts a second run.

Input parameters:

- `cwd` (string)
- `effort` (string, required)
- `idempotencyKey` (string, required)
- `limits` (object)
- `model` (string, required)
- `prompt` (string, required)
- `provider` (string, required)
- `serviceTier` (string)
- `workspaceId` (string, required)

### `continueAgentSession` (~76 tokens)

Send a follow-up prompt to the conversation of a finished run, using its sessionId. Keeps the same agent, model, workspace and folder. Returns a new taskId.

Input parameters:

- `idempotencyKey` (string, required)
- `limits` (object)
- `prompt` (string, required)
- `sessionId` (string, required)

### `getAgentRunStatus` (~44 tokens)

Get a run's state, timing, token usage and error. succeeded means the agent finished, not that its work is correct.

Input parameters:

- `taskId` (string, required)

### `getAgentRunOutput` (~57 tokens)

Read a run's messages, commands, tool calls and file changes after cursor, plus the final response once it has finished.

Input parameters:

- `cursor` (integer)
- `limit` (integer)
- `taskId` (string, required)

### `listAgentRuns` (~48 tokens)

List runs, optionally filtered by workspace and status.

Input parameters:

- `cursor` (string)
- `limit` (integer)
- `status` (string)
- `workspaceId` (string)

### `cancelAgentRun` (~32 tokens)

Stop a queued or running run. Changes it already made are not undone.

Input parameters:

- `taskId` (string, required)

## Diagnostics

Captured diagnostic sections: Provenance, Dependencies. The full working is on the page: https://verifymcp.io/servers/nishkalkashyap-agentgo/agentgo-mcp#diagnostics

## Score history

- 2026-10-17: 65
- 2026-10-16: 65
- 2026-10-15: 65
- 2026-10-14: 65
- 2026-10-13: 39
- 2026-10-12: 39
- 2026-10-11: 24
- 2026-10-09: 39

## Common questions

### What is the AgentGo MCP server?

AgentGo is an MCP server listed in the public MCP registry as io.github.Nishkalkashyap/agentgo. Run Codex and Claude Code on your computer from an MCP client, and track or continue their runs. This page covers its npm package (agentgo-mcp).

### Is the AgentGo MCP server safe to use?

AgentGo scores 65 out of 100 on VerifyMCP. We found no known CVEs affecting it as of 17 October 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 AgentGo MCP server expose?

AgentGo exposes 13 tools: getAgentCapabilities, getSupportedModels, listWorkspaces, listDirectory, readFile, and 8 more. Their descriptions and schemas cost roughly 892 tokens of context every time the server is loaded.

### Is the AgentGo MCP server still maintained?

AgentGo is still listed as active in the MCP registry. We last reached this channel on 17 October 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced.

### What licence is the AgentGo MCP server under?

AgentGo declares the MIT licence, which is OSI-approved. That covers the source only, and says nothing about the cost of any service it calls.

## Links

- npm package: https://www.npmjs.com/package/agentgo-mcp
- Socket report: https://socket.dev/npm/package/agentgo-mcp
- Repository: https://github.com/Nishkalkashyap/agentgo
- Changelog RSS feed: https://verifymcp.io/servers/nishkalkashyap-agentgo/agentgo-mcp.xml
- Changelog JSON feed: https://verifymcp.io/servers/nishkalkashyap-agentgo/agentgo-mcp.json
- HTML version of this page: https://verifymcp.io/servers/nishkalkashyap-agentgo/agentgo-mcp