momentic-spec Skill for Claude Code: What It Does and How to Install It

Makes Claude write the Momentic tests for a feature first, then implement the feature, then run the tests at a checkpoint.

Installs: 28.3K; on skills.sh
Stars: 16; on GitHub
Files: 1; besides SKILL.md
Updated: Oct 2026; last repo commit

This skill makes Claude treat Momentic tests as the product spec. It follows a fixed 7-step sequence: read the repo, report the affected tests, write or change them, review the test diff, implement the feature, use fast code checks, then ask before running Momentic at a durable checkpoint.

It hands detailed test authoring to the momentic-test skill, so install both. It needs a Momentic project and sign-in.

Good for

Developers who want a feature spec as Momentic tests before the code
Teams that want Claude to stop changing tests to fit its own implementation

Install momentic-spec

Claude Code

Copy this prompt.

Install the momentic-spec skill in Claude Code. Run this command:

/plugin marketplace add momentic-ai/skills
/plugin install momentic@momentic

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

The plugin route installs all 6 skills (skills/*/SKILL.md in the repo tree) and registers the momentic and momentic-mobile MCP servers from .mcp.json. SKILL.md tells Claude to use the installed momentic-test skill, so every route below installs momentic-spec and momentic-test together. The skill edits tests and product code in your repo, so there is no Claude app upload route. Sign in once with the login command in the last method, or set MOMENTIC_API_KEY.

From the project README. Source.

What's in the skill

When it loads

Improve code correctness using Momentic specs in the feature development process

Sections in SKILL.md

Non-negotiable sequence
Choose the affected tests
Write the product specification first
Follow the codebase's testing language
Prefer AI action V3 for new specifications
Run at durable end-to-end checkpoints
Own the test data and application state
Edit and validate safely

Key facts

Author: Momentic (vendor's own repo)
License: None stated: the repo has no LICENSE file and its plugin manifests list no license
Installs on skills.sh: 28,277 (Nov 2, 2026)
GitHub stars: 16 for the whole repo (Nov 2, 2026)
Last repo commit: Oct 28, 2026
Audits on skills.sh: Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
What it includes: 2 files: SKILL.md and agents/openai.yaml (from the repo tree)
Needs: The momentic-test skill, a Momentic project, a Momentic sign-in, and the Momentic MCP server
Part of: The Momentic plugin: 6 skills (skills/*/SKILL.md in the repo tree) and 2 local MCP servers listed in .mcp.json
Upstream sync: The README says a bot copies most files in skills/ from an upstream source, so edits made in the repo are overwritten

What it includes

Read repo, name affected tests, write tests, review, implement, check, run
Product code does not start until the test specs are on disk.
Try: “Use /momentic-spec to define the affected Momentic tests first, then implement this feature.”
Write · Seven-step sequence
Update old-behavior tests; add new ones only for new journeys
It reports each test path and whether it is added, changed or unchanged before editing product code.
Read · Choose the affected tests
Tests describe the outcome, even if the product cannot pass yet
Not-yet-passing tests get disabled: true, with the reason and what must become true.
Write · Write the spec first
Reuse modules, nearby conventions and native Momentic steps
It never works around a real product failure by changing the test.
Read · Follow the codebase style
Use act: steps for long flows; keep cache on for stable flows
Set cache: false only for inherently dynamic goals.
Prefer AI action V3
Propose the smallest test set and ask before a long run
Confirmation covers one long run; it asks again before another unless you allowed reruns.
Runs code · Run at checkpoints
Seed data, accounts and flags before the Momentic run
It may add test-only endpoints or scripts, which must be gated to test environments.
Write · Own the test data
Affected tests, enabled or disabled state, behavior built, checks run
Lists every validation that was or was not run.
Handoff report

From the official docs

Permissions

Medium risk

This skill writes both tests and product code in your repo, and may add test-only endpoints or scripts. Review the test diff first, and check that anything it adds for test data cannot run in production.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

It writes product code, not only tests
After the tests are on disk, the sequence has Claude implement the product change and run the repository checks. Review both diffs. github.com
It may add test-only endpoints or scripts
SKILL.md encourages seed endpoints, fixtures, test accounts and feature-flag controls when none exist. It says they must be inaccessible in production, gated to approved test environments and safe to call repeatedly. github.com
Long Momentic runs need your yes
SKILL.md says to stop before a long-running run, name the tests, and run only after you confirm. A direct request to run specific tests counts as confirmation. github.com
The plugin route adds two local MCP servers
The plugin route installs the repo's .mcp.json, which starts momentic and momentic-mobile with npx -y. The repo tree has no hooks. Two skills (mo-qa and momentic-spec) bundle an agents/openai.yaml that holds a display name, a short description and a default prompt. github.com
It ships interface metadata
agents/openai.yaml sets a display name, a short description and a default prompt. It holds no commands. github.com

Security advisories

No published security advisories as of Nov 2, 2026.

Technical details

Turning on read-only	SKILL.md describes no read-only mode.
Talks to	Momentic MCP and CLI services used when tests run
Sign-in and keys	Sign in once with npx momentic login, or set MOMENTIC_API_KEY (README). SKILL.md does not say where the login is stored.
Scopes	Reads your repo guidance, momentic.config.yaml and nearby tests Writes Momentic test files and your product code May add test-only seed endpoints, scripts or fixtures Runs Momentic tests after you confirm a long run

Questions

What does the momentic-spec skill do?

It makes Claude write the expected Momentic tests for a feature before it writes the product code. It then implements the change and, at a durable checkpoint, asks whether to run the smallest affected set of tests.

How do I install it?

Use the install box at the top of this page. Every route there installs momentic-spec together with momentic-test, because SKILL.md tells Claude to use the installed momentic-test skill for detailed authoring. The plugin route installs all six Momentic skills.

Do I need a Momentic account?

It needs a Momentic project and sign-in, because the tests it writes are Momentic tests. The README says to run npx momentic login once or set MOMENTIC_API_KEY.

Is it an official Momentic skill?

Yes. It lives in Momentic's own repo, momentic-ai/skills, and the plugin manifest names Momentic as author. The README says a bot copies most skill files from an upstream source, so Momentic asks outside contributors to open an issue instead of a pull request.

Is it safe?

It writes tests and product code in your repo and may add test-only seed endpoints or scripts, which SKILL.md says must not be reachable in production. It asks before a long Momentic run. Review the test diff before the code is written.

Does it work in Cursor or Codex?

The README documents a Cursor plugin from the Cursor Marketplace and a Devin plugin. It also says install-mcp writes the MCP server entry for Claude Code, Cursor, VS Code, Codex, Windsurf and other clients. The skill files are plain SKILL.md, so the skills CLI can install them in other agents.

Why does it need momentic-test?

SKILL.md says to use the installed momentic-test skill for detailed authoring, format and MCP guidance, and links to its time-sensitive interactions section.

What does disabled: true do here?

SKILL.md marks a spec test disabled when the implementation or a dependency is not ready and the test is not expected to pass yet. It says to record why, and to remove the flag at the checkpoint when the test should pass.

What prompt starts it?

The skill's agents/openai.yaml sets this default prompt: "Use /momentic-spec to define the affected Momentic tests first, then implement this feature."

Information

updated Nov 2, 2026

Repository: momentic-ai/skills
Folder: skills/momentic-spec

momentic-ai/skills skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 6

README · github.com
.mcp.json · github.com
marketplace.json · github.com
momentic-spec SKILL.md · github.com
momentic-test/SKILL.md · github.com
momentic-spec openai.yaml · github.com