{
  "markdown": "# Momentic Agent Skills\n\nAgent skills for [Momentic](https://momentic.ai), an end-to-end testing platform\nfor web and mobile apps.\n\nThe skills teach a coding agent how to write Momentic tests, run them, and fix\nthem when they break. The package also wires up the local Momentic MCP servers,\nwhich are what let the agent drive a real browser or device while it works.\n\n## Installation\n\n### Claude Code\n\nAdd this repository as a Claude Code marketplace, then install the Momentic\nplugin:\n\n```shell\n/plugin marketplace add momentic-ai/skills\n/plugin install momentic@momentic\n```\n\nThe plugin installs all the skills and registers the local Momentic MCP servers.\nRun `/mcp` in Claude Code to see which ones started.\n\n### Cursor\n\nInstall the Momentic plugin from the Cursor Marketplace. Search for\n**Momentic** under **Customize → Plugins**.\n\n### Devin\n\nInstall the Devin plugin from this repository. It includes all the skills and\nthe local Momentic MCP servers.\n\n### Other agents\n\nInstall the skills with the `skills` CLI:\n\n```shell\nnpx skills add momentic-ai/skills\n```\n\nThis repository is also an Agent Plugins v1 package. The manifest is at\n[`plugin.json`](plugin.json).\n\n## MCP servers\n\nThe skills call MCP tools that run tests, inspect live pages, and splice steps\nback into your YAML files. Those tools come from the Momentic CLIs over stdio, so\nthe servers run on your machine next to your project:\n\n```json\n{\n  \"mcpServers\": {\n    \"momentic\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"momentic\", \"mcp\"]\n    },\n    \"momentic-mobile\": {\n      \"command\": \"npx\",\n      \"args\": [\"-y\", \"momentic-mobile\", \"mcp\"]\n    }\n  }\n}\n```\n\nThat is what `.mcp.json` and the Devin manifest install. `momentic` drives\nbrowsers and `momentic-mobile` drives Android and iOS, so drop whichever one\nyour project does not test.\n\nBoth servers look for `momentic.config.yaml` in the working directory and its\nparents. Add `--config /absolute/path/to/momentic.config.yaml` if your agent\nstarts outside the project, or if the two platforms live in separate\nsubdirectories.\n\nSign in once with `npx @momentic/wizard@latest login`, or set\n`MOMENTIC_API_KEY` in the environment. The wizard can also register a server\nwith Claude Code, Cursor, VS Code, Codex, Windsurf, and a few others if you\nwould rather not edit config by hand. Mobile work needs the usual device\ntoolchain, and `momentic-mobile mcp` takes `--android-home` and `--java-home`\nwhen those live somewhere unusual. See the\n[MCP server documentation](https://momentic.ai/docs/coding-agents/mcp-server)\nfor the full tool list.\n\n## Skills\n\nEach skill is a Markdown file that gives an agent domain knowledge and a\nstep-by-step workflow. They use the\n[`SKILL.md` format](https://docs.cursor.com/context/rules-for-ai).\n\n| Skill                                                                            | Description                                                                                                                                                                                                                                                                                                                                                                          |\n| -------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |\n| [momentic-test](skills/momentic-test/SKILL.md)                                   | Create, run, and maintain Momentic browser E2E tests and modules stored as `*.test.yaml` and `*.module.yaml` files.                                                                                                                                                                                                                                                                  |\n| [momentic-mobile-test](skills/momentic-mobile-test/SKILL.md)                     | Create, run, and maintain Momentic mobile E2E tests and modules for Android and iOS. Use Momentic MCP tools for live device validation, and use direct v2 YAML edits only for high-confidence local mobile v2 changes.                                                                                                                                                               |\n| [momentic-result-classification](skills/momentic-result-classification/SKILL.md) | Classify or explain Momentic test run results using Momentic MCP tools. Use when the user asks to categorize a failure, understand why a run failed, triage test results, or compare run results to past run results.                                                                                                                                                                |\n| [momentic-maintain](skills/momentic-maintain/SKILL.md)                           | Diagnose, classify, triage, and repair failing Momentic tests with MCP run tools, the Momentic CLI, and manual run artifacts. Use when a developer asks what happened on a branch, DevX or on-call asks why main is red, or the user wants to inspect classifications, de-flake quarantined or recovered tests, reduce retries, re-classify runs, run AI triage, or repair failures. |\n| [momentic-spec](skills/momentic-spec/SKILL.md)                                   | Improve code correctness using Momentic specs in the feature development process                                                                                                                                                                                                                                                                                                     |\n\n## Skill sources\n\nA bot copies most files in `skills/` from an upstream source, so anything you\nedit here gets overwritten on the next sync. That covers `momentic-test`,\n`momentic-mobile-test`, `momentic-maintain`, `momentic-spec`, and\n`momentic-triage-quarantined-tests.md`. If you work at Momentic, edit them at\nthe source. If you do not, please open an issue instead of a pull request.\n",
  "bytes": 6366,
  "sha": "d2787a371f31b519193ff71f6a335b536e561e43a954f7d0000ada3f238a6916",
  "repo_slug": "momentic-ai/skills",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/skl_momentic_ai_skills_momentic_maintain_712748e9/readme"
}