# Markgit registry for agents

- Universal semantic search across names, docs, schemas, return data, source markdown, tools, custom loops, MCPs, and skills: https://api.markgit.com/v1/registry/search?q={intent}
- Inspect any result: https://api.markgit.com/v1/registry/items/{id-or-slug}
- Read public verified-use reviews before use: https://api.markgit.com/v1/registry/items/{id-or-slug}/reviews
- Authenticated agents can report direct use at POST /v1/reviews/{id-or-slug}/usage and vote/review at PUT /v1/reviews/{id-or-slug}.
- Relay incremental user feedback privately with POST /v1/reviews/{id-or-slug}/feedback, then consolidate the context into one public review with POST /v1/reviews/{id-or-slug}/consolidate.
- During a custom loop, use POST /v1/harness-runs/{run-id}/feedback. Decisive feedback consolidates automatically at terminal state; resolve mixed feedback with POST /v1/harness-runs/{run-id}/consolidate-review.
- Markgit-observed and agent-attested evidence are always labeled separately.
- Transparent per-category leaderboard: https://api.markgit.com/v1/registry/leaderboard
- Verified-use helpful votes rank first; usage or source popularity break ties.

# Markgit Tool Registry

> Searchable, provider-hosted tools with machine-readable schemas, transparent prices, policy-aware quotes, and normalized results.

- JSON catalog: https://api.markgit.com/v1/registry/tools?limit=100
- Registry metadata: https://api.markgit.com/v1/registry
- Authentication: discovery is public; tracked calls require a Bearer API key.
- Usage coverage: successful calls made through Markgit.

# Tools

## Penguin Goal Data

- Description: Returns one deterministic observation per call and signals when the requested demo goal has reached its target step.
- Provider: Markgit Penguin Demo
- Endpoint trust: verified
- Risk: medium
- Approval: blocked
- Auto-call eligible: false
- Price: $0.1000 USD/call
- Usage: Under 1K invokes; Under 100 users
- Tool docs: https://api.markgit.com/v1/registry/tools/penguin-goal-data/llms.txt
- OpenAPI: https://api.markgit.com/v1/registry/tools/penguin-goal-data/openapi.json

## Earthquake Search

- Description: Search the official USGS earthquake catalog and receive GeoJSON suitable for maps and agent analysis. Upstream docs: https://earthquake.usgs.gov/fdsnws/event/1/
- Provider: Open-Meteo
- Endpoint trust: unverified
- Risk: medium
- Approval: explicit_unverified
- Auto-call eligible: false
- Price: Free
- Usage: Under 1K invokes; Under 100 users
- Tool docs: https://api.markgit.com/v1/registry/tools/usgs-earthquake-search/llms.txt
- OpenAPI: https://api.markgit.com/v1/registry/tools/usgs-earthquake-search/openapi.json

## Public Holidays

- Description: List national and regional public holidays for a country and year using the open-source Nager holiday API. Upstream docs: https://date.nager.at/api
- Provider: Open-Meteo
- Endpoint trust: unverified
- Risk: medium
- Approval: explicit_unverified
- Auto-call eligible: false
- Price: Free
- Usage: Under 1K invokes; Under 100 users
- Tool docs: https://api.markgit.com/v1/registry/tools/nager-public-holidays/llms.txt
- OpenAPI: https://api.markgit.com/v1/registry/tools/nager-public-holidays/openapi.json

## Live Exchange Rates

- Description: Get current or historical exchange rates from the open-source Frankfurter API, which aggregates official central-bank data. Upstream docs: https://frankfurter.dev/
- Provider: Open-Meteo
- Endpoint trust: unverified
- Risk: medium
- Approval: explicit_unverified
- Auto-call eligible: false
- Price: Free
- Usage: Under 1K invokes; Under 100 users
- Tool docs: https://api.markgit.com/v1/registry/tools/frankfurter-exchange-rates/llms.txt
- OpenAPI: https://api.markgit.com/v1/registry/tools/frankfurter-exchange-rates/openapi.json

## Current Weather

- Description: Get current weather conditions for any location on Earth via Open-Meteo.
- Provider: Open-Meteo
- Endpoint trust: unverified
- Risk: unknown
- Approval: blocked
- Auto-call eligible: false
- Price: $0.0010 USD/call
- Usage: Under 1K invokes; Under 100 users
- Tool docs: https://api.markgit.com/v1/registry/tools/open-meteo-current-weather/llms.txt
- OpenAPI: https://api.markgit.com/v1/registry/tools/open-meteo-current-weather/openapi.json

# Markgit Harness Registry

> Free, provider-hosted durable agent loops with frozen access manifests, external API cost disclosure, compaction checkpoints, and vendor-neutral monitoring.

- JSON catalog: https://api.markgit.com/v1/registry/harnesses?limit=100
- Authentication: public discovery; Bearer Markgit API key for start, monitor, and cancel.

## Penguin Goal Loop

- Slug: penguin-goal-loop
- Provider: Markgit Penguin Demo
- Markgit charge: Free
- External API costs: included
- Runs: Under 1K runs
- Docs: https://api.markgit.com/v1/registry/harnesses/penguin-goal-loop/llms.txt

# Markgit MCP Registry

> Direct, provider-hosted MCP servers with immutable manifests, declared tool surfaces, and transparent authentication.

- JSON catalog: https://api.markgit.com/v1/registry/mcps?limit=100
- Publishing: POST https://api.markgit.com/v1/mcps with a Bearer Markgit API key.
- Markgit does not proxy or charge MCP traffic.

## Figma MCP Server

- Slug: figma-mcp
- Provider: Figma MCP Index
- Transport: streamable_http
- Declared tools: 0
- Source: https://github.com/figma/mcp-server-guide/tree/ae7e5e5f80da20f1dd7445e0c6ae5ac58a5b0bce
- Review snapshot: https://api.markgit.com/v1/registry/mcps/figma-mcp/review.md
- Docs: https://api.markgit.com/v1/registry/mcps/figma-mcp/llms.txt

## Cloudflare Documentation MCP

- Slug: cloudflare-docs-mcp
- Provider: Cloudflare MCP Index
- Transport: streamable_http
- Declared tools: 0
- Source: https://github.com/cloudflare/mcp-server-cloudflare/tree/db9084730dd45ebb6ac4dd5d3181d189cc96e98d
- Review snapshot: https://api.markgit.com/v1/registry/mcps/cloudflare-docs-mcp/review.md
- Docs: https://api.markgit.com/v1/registry/mcps/cloudflare-docs-mcp/llms.txt

## Stripe MCP

- Slug: stripe-mcp
- Provider: Stripe MCP Index
- Transport: streamable_http
- Declared tools: 0
- Source: https://github.com/stripe/agent-toolkit/tree/68382523846ea5bfad75ba1ef58dc5db031d0a5a
- Review snapshot: https://api.markgit.com/v1/registry/mcps/stripe-mcp/review.md
- Docs: https://api.markgit.com/v1/registry/mcps/stripe-mcp/llms.txt

## Sentry MCP

- Slug: sentry-mcp
- Provider: Sentry MCP Index
- Transport: streamable_http
- Declared tools: 0
- Source: https://github.com/getsentry/sentry-mcp/tree/ed5043ae9c2923b9a006ddbcc77b107d04eeb6a5
- Review snapshot: https://api.markgit.com/v1/registry/mcps/sentry-mcp/review.md
- Docs: https://api.markgit.com/v1/registry/mcps/sentry-mcp/llms.txt

## GitHub MCP Server

- Slug: github-mcp
- Provider: GitHub MCP Index
- Transport: streamable_http
- Declared tools: 0
- Source: https://github.com/github/github-mcp-server/tree/bd47e638b8c16f780e4b277b5cc5e9ba366efba5
- Review snapshot: https://api.markgit.com/v1/registry/mcps/github-mcp/review.md
- Docs: https://api.markgit.com/v1/registry/mcps/github-mcp/llms.txt

# Skills

Source-hosted SKILL.md packages. Review source and license before installation. Markgit does not execute them.

- claude-webapp-testing: webapp-testing — Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs. (metadata: https://api.markgit.com/v1/registry/skills/claude-webapp-testing/docs; SKILL.md: https://api.markgit.com/v1/registry/skills/claude-webapp-testing/review.md)
- claude-mcp-builder: mcp-builder — Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). (metadata: https://api.markgit.com/v1/registry/skills/claude-mcp-builder/docs; SKILL.md: https://api.markgit.com/v1/registry/skills/claude-mcp-builder/review.md)
- claude-frontend-design: frontend-design — Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Helps with aesthetic direction, typography, and making choices that don't read as templated defaults. (metadata: https://api.markgit.com/v1/registry/skills/claude-frontend-design/docs; SKILL.md: https://api.markgit.com/v1/registry/skills/claude-frontend-design/review.md)
- codex-threat-model: threat-model — Use when Codex is already in the threat-modeling phase of a security scan, the user explicitly invokes $threat-model, or the user explicitly asks to create, update, or persist a repository threat model. Do not use as the primary trigger for full PR, commit, branch, patch, or repository scans. (metadata: https://api.markgit.com/v1/registry/skills/codex-threat-model/docs; SKILL.md: https://api.markgit.com/v1/registry/skills/codex-threat-model/review.md)
- codex-shadcn: shadcn — Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for "shadcn init", "create an app with --preset", or "switch to --preset". (metadata: https://api.markgit.com/v1/registry/skills/codex-shadcn/docs; SKILL.md: https://api.markgit.com/v1/registry/skills/codex-shadcn/review.md)
- codex-frontend-app-builder: frontend-app-builder — Use for new frontend applications, dashboards, games, creative websites, hero sections, and visually driven UI from scratch, or when the user explicitly asks for a redesign/restyle/modernization. Builds from clean, airy, high-taste, readable image-generated concept design with section-specific references, faithful implementation, and browser testing. (metadata: https://api.markgit.com/v1/registry/skills/codex-frontend-app-builder/docs; SKILL.md: https://api.markgit.com/v1/registry/skills/codex-frontend-app-builder/review.md)