Log inSubmit server
Agentry Observability logo

Agentry Observability

Provides an MCP handoff that guides coding agents to add Agentry analytics, error logging, and deploy attribution.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Agentry Observability

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "agentry-observability": {
      "command": "npx",
      "args": [
        "-y",
        "@agentrysh/mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

We couldn’t confirm it starts correctly

The install command started, but didn't respond the way we expected when we tried to talk to it.

npx -y @agentrysh/mcp

initialize succeeded but no response to tools/list.

This is an experimental automated check and can have false negatives — missing environment variables, a slow cold install, etc. It doesn’t necessarily mean something’s wrong. Last checked 15d ago.

Overview

Agentry Observability MCP server gives AI coding agents the canonical Agentry skill handoff for adding analytics, error logging, and deploy attribution. It runs locally through the `@agentrysh/mcp` package and returns guidance for working with Agentry's HTTP API. The MCP layer does not ingest telemetry or proxy API requests, so use it when an agent needs implementation guidance rather than a runtime observability pipeline.

Use cases

  • Add product analytics to an application
  • Instrument exception and failed-job logging
  • Track releases and code changes in production
  • Guide an agent through Agentry installation
  • Add an agent-readable observability layer

Key features

  • Returns the canonical Agentry skill handoff
  • Guides product and business-event analytics
  • Covers exception and operational-failure logging
  • Documents deploy attribution
  • Supports agent-led setup verification
  • Links to Agentry API and OpenAPI resources

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 3d ago

Recent health check succeeded.

7/7 checks healthy over the last 41d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 16d ago via OSV.dev · @agentrysh/mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars2Stargazers on the source repository.
npm downloads461Package downloads in the last 30 days.
Last commit3mo agoMost recent push to the default branch.
Install checkInconclusiveStarted but did not complete the MCP handshake in our test.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How Agentry Observability works

What Agentry Observability MCP server does

The Agentry Observability MCP server exposes the canonical Agentry skill handoff to AI coding agents through MCP. The handoff explains how to add three related observability capabilities: product analytics, error logging, and deploy attribution.

Product analytics covers events related to users, funnels, activation, retention, and business activity. Error logging addresses exceptions, failed jobs, operational failures, and supporting case context. Deploy attribution links production behavior with releases and code changes, helping an agent connect an observed issue or behavior to recent changes.

The repository is a discovery and distribution surface for Agentry across MCP, Claude Code, Codex, and agent IDE adapter ecosystems. It also contains shared Claude Code and Codex plugin material, marketplace catalogs, and links to the live adapter manifest.

How it works

The Agentry Observability MCP server is started as a local standard-input/standard-output process with npx. An MCP-compatible client can then request the Agentry skill handoff while working in a codebase. That handoff is intended to help an agent install and configure Agentry from the actual project, with verification before setup is considered complete.

The MCP package is not an observability collector. Runtime applications write directly to Agentry's HTTP API. The MCP server and skill do not proxy that API, ingest telemetry, or replace the live documentation. For API discovery and schemas, the project points to Agentry's API site and OpenAPI document.

Reach for the Agentry Observability MCP server when an agent needs to implement analytics, error monitoring, product telemetry, deploy tracking, production debugging, or an agent-readable observability layer. It is less suitable when you need MCP tools that query existing telemetry or perform monitoring operations directly.

Setup and configuration

Run the package with:

Terminal
npx -y @agentrysh/mcp

A compatible client can configure it as a server named agentry, using npx as the command and -y, @agentrysh/mcp as its arguments. The provided setup does not specify environment variables, API keys, or additional MCP configuration values.

Agentry also distributes related instructions outside MCP. Claude Code can install the skill under ~/.claude/skills/agentry, and the repository provides marketplace plugins for Claude Code and Codex. IDE and agent-specific instruction paths are published through the live adapter manifest.

Tools and capabilities

The documented MCP surface returns the canonical Agentry skill handoff. Its guidance covers:

  • Product, funnel, activation, retention, and business-event analytics.
  • Exception, failed-job, operational-failure, and case-context logging.
  • Release and code-change attribution for production behavior.
  • Agent-led installation and verification from the target codebase.
  • Links to the live skill, installation documentation, API discovery endpoint, and OpenAPI schema.

The repository also advertises support across multiple agent surfaces, including Codex, Claude Code, Cursor, VS Code with GitHub Copilot, Windsurf Cascade, Cline, Roo Code, Continue, Zed, Gemini CLI, generic MCP clients, and AGENTS.md-aware agents.

Limitations and notes

The Agentry Observability MCP server supplies implementation guidance, not direct monitoring access. It does not ingest application events, query telemetry, send logs, or replace Agentry's HTTP API. Applications must write runtime data directly to that API.

The repository describes the MCP package as a handoff surface rather than a collection of separately documented analytics or logging operations. Developers looking for live data inspection should use the Agentry API or another purpose-built observability interface instead.

Client adapter support can change independently of the repository's static listings. The live adapter manifest is the canonical source for current native instruction-file paths. Codex marketplace availability may also depend on adding the Git-backed marketplace source or receiving workspace access.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Agentry Observability

Agentry Observability is an MCP server that connects AI coding agents to the Agentry observability workflow through a canonical skill handoff. Its main capability is providing guidance for product analytics, error logging, and deploy attribution. It does not proxy Agentry's HTTP API, ingest telemetry, or query runtime monitoring data directly.

Run `npx -y @agentrysh/mcp` and register that command in an MCP-compatible client. The documented configuration uses `npx` with `-y` and `@agentrysh/mcp` as arguments, under a server name such as `agentry`.

Yes, Cursor is listed among Agentry's supported agent surfaces. The MCP package itself uses a standard local command configuration, while the live adapter manifest provides the current native instruction-file path for supported clients.

No, the MCP server does not query production telemetry. It returns the canonical Agentry skill handoff and related guidance; runtime applications write directly to Agentry's HTTP API.

Related MCP servers

All in Monitoring All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewAgentry Observability AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/agentry-observability?style=directory)](https://allmcps.com/mcp/agentry-observability)
HTML Embed
<a href="https://allmcps.com/mcp/agentry-observability"><img src="https://allmcps.com/api/badge/agentry-observability?style=directory" alt="Agentry Observability on AllMCPs" /></a>