Not in the top 1,000 by GitHub stars

PyModel/niblet-skill-mcp

MCP server and design skill that grounds coding-agent UI work in real product screens. Two tools, npx-installable.

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

skill/niblet/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

cf81b1b8ca03 → cf81b1b8ca03

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → cf81b1b8ca03

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit cf81b1b8ca03af367082fc7e781551bd79c94ad1

git clone --filter=blob:none --no-checkout https://github.com/PyModel/niblet-skill-mcp.git .hotskills/pymodel-niblet-skill-mcp && git -C .hotskills/pymodel-niblet-skill-mcp checkout --detach cf81b1b8ca03af367082fc7e781551bd79c94ad1 # Copy each selected skill directory from .hotskills/pymodel-niblet-skill-mcp into .agents/skills/<skill-name>/
74GitHub stars
-15stars in 7 days
1skill in this repository
6commits in 30 days

Heat

1.2
  • Star growth, last 7 daysNot available
  • Star growth, last 24 hoursNot available
  • Rank movement, last 7 daysNot available
  • Commits, last 30 daysNot available
  • Accounts asking about it through MCP, last 7 daysNot available
  • Hot votes this weekNot available

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Oct 25, 2026Nov 5, 202696

30-day change not yet available: not enough history.

Individual skills in this repository

How it works

niblet

When your agent uses it

Keep interface work anchored to the product it belongs to instead of a generic template. Sets a short design contract, builds from the components and tokens already in the codebase, covers the states a surface can actually reach, and closes by looking at the rendered result. Use when building, reworking, or assessing a web or native interface. Trigger with "niblet", "niblet skill", "niblet designer ui", or "niblet review". Skip backend, CLI, data, and infrastructure work, prose-only tasks, and questions the product's own design system already settles.

What it loads into context

Frontmatter

239 tokens

Always available to the agent

SKILL.md

2,230 tokens

When the skill is activated

Reference documents

7,534 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Start here
  2. Four modes
  3. Design contract
  4. Slop checks
  5. Required state coverage
  6. Bounded rendered finish gate
  7. Optional connection and helpers

Files

  • LICENSEasset10,791 bytes
  • NOTICEasset183 bytes
  • SKILL.mdcore11,798 bytes
  • SKILL.mdcore11,798 bytes
  • SKILL.mdcore11,798 bytes
  • agents/openai.yamlasset441 bytes
  • references/commands.mddoc11,636 bytes
  • references/connection.mddoc16,105 bytes
  • references/evidence.mddoc7,969 bytes
  • references/native.mddoc3,160 bytes
Read SKILL.md

# Niblet

Build an interface that belongs to this product, not a generic template. [Niblet](https://niblet.pymodel.com) is an optional reference catalogue; this skill also works entirely from local product evidence without an MCP connection, account, script, or browser automation.

## Start here

1. Identify the requested surface, action, and scope. Read the product brief, relevant current screens, components, tokens, content, and platform constraints. Inspect the repository directly; this skill has no setup script.
2. Select the surface mode below and write a compact design contract in the conversation or the existing surface brief. Ask only for a consequential decision that the available product evidence cannot answer. Mark other assumptions.
3. Read [the command playbook](references/commands.md) for the requested command. The local MCP adapter registers each entry as `niblet-<command>`; a standalone filesystem installation invokes the same work instructions in ordinary language. For native work, also read [native guidance](references/native.md).
4. Work within the contract. Use existing components and tokens before introducing new ones. If a concrete visual question remains unresolved, follow [the evidence policy](references/evidence.md); reference retrieval is optional, never a prerequisite to useful work.
5. Cover the applicable states and run the bounded finish gate. Report the delivered scope, evidence actually inspected, and any specific verification limitation.

**Routing:** with no target or command, present a short context-aware menu and wait for a choice rather than making changes. An explicit command loads its playbook entry. Otherwise treat the request as ordinary design work. For a new surface or replacement visual world without product context, run `init` first; narrow refinements can proceed from the incumbent implementation. `craft` aliases ordinary new work; `teach` aliases `init`.
View the complete source on GitHub

Skills included

  • niblet

Repository facts

License
Apache-2.0
Created
Sep 6, 2026
Last pushed
Oct 17, 2026
Forks
4
Open issues
0
Project activity
Active

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 4, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 2 notes
  • skill/niblet/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skill/niblet/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 26 signals by file

skill/niblet/LICENSE

  • Network accessline 1
    Apache License                            Version 2.0, January 2004                         http://www.apache.org/licenses/
  • Network accessline 3
    http://www.apache.org/licenses/
  • Network accessline 185
    http://www.apache.org/licenses/LICENSE-2.0

skill/niblet/NOTICE

  • Network accessline 6
    https://github.com/PyModel/niblet-skill-mcp

skill/niblet/SKILL.md

  • Network accessline 7
    author: 'Mohamed Elkholy (elkaix)'   organization: 'PyModel'   source: 'https://github.com/PyModel/niblet-skill-mcp'
  • Network accessline 8
    organization: 'PyModel'   source: 'https://github.com/PyModel/niblet-skill-mcp'   compatibility: 'Claude Code, Codex, Cursor, and GitHub Copilot, or any agent that can open repository files and call an MCP tool.'
  • Network accessline 9
    source: 'https://github.com/PyModel/niblet-skill-mcp'
  • Network accessline 16
    Build an interface that belongs to this product, not a generic template. [Niblet](https://niblet.pymodel.com) is an optional reference catalogue; this skill also works entirely from local product evidence without an MCP connection, account,

skill/niblet/agents/openai.yaml

  • Network accessline 5
    author: "Mohamed Elkholy (elkaix)"   organization: "PyModel"   source: "https://github.com/PyModel/niblet-skill-mcp"
  • Network accessline 6
    organization: "PyModel"   source: "https://github.com/PyModel/niblet-skill-mcp"   license: "Apache-2.0"
  • Network accessline 7
    source: "https://github.com/PyModel/niblet-skill-mcp"

skill/niblet/references/connection.md

  • Network accessline 7
    The skill can work from the repository, product brief, and supplied screenshots without MCP. The catalogue site is [https://niblet.pymodel.com](https://niblet.pymodel.com); browsing it needs a signed-in account with access. Browser inspecti
  • Network accessline 14
    | ----------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------- | --------------------------------------------------
  • Network accessline 15
    | MCP `find_ui_references` / `find_ui_materials` / `get_design_reference`                                           | Preferred agent evidence path                                         | 1–3 results per call                              
  • Network accessline 16
    [Credential-related source omitted]
  • Network accessline 22
    GET https://niblet-api.pymodel.com/v1/search?q=<question>&limit=3
  • Network accessline 26
    To inspect a returned screen, call `GET https://niblet-api.pymodel.com/v1/screens/<id>` with the same header. Read only the JSON payload, choose at most three IDs, and stop. Do not paginate the catalogue or raise `limit` above 3 for agent w
  • Network accessline 28
    Without a key there is no catalogue fallback: every catalogue route answers 401. Continue from local product evidence, and tell the user that references need a Niblet account and a key from [niblet.pymodel.com/account](https://niblet.pymode
  • Network accessline 32
    Prerequisite: Node.js **24.15 or later**. The adapter's bundled documents, `niblet_help`, and `niblet_status` work without a token. Its three catalogue tools call REST `/v1`. A public `niblet_at_…` account key created at [niblet.pymodel.com
  • Network accessline 55
    Niblet itself is a hosted service — there is nothing to self-host. For catalogue access, point this adapter at the hosted API: set `NIBLET_TOKEN` to an account key created at `https://niblet.pymodel.com/account` (one key works on the REST c
  • Network accessline 57
    The package contacts `https://niblet-api.pymodel.com` by default, or the HTTP(S) origin in `NIBLET_API_ORIGIN`. It sends the token as bearer authentication for API requests, and never to the media origin. `NIBLET_TOKEN` configures this loca
  • Network accessline 61
    If a catalogue tool fails, read the error: it tells you whether to relay a config change to the user (missing, unexpanded, or placeholder token; website origin instead of `niblet-api.pymodel.com`) or to continue from `niblet://skill` withou
  • Network accessline 101
    Images arrive as MCP image content, fetched only from the configured media origin (`https://niblet.pymodel.com` by default, or `NIBLET_MEDIA_ORIGIN`) and the API origin, without the token, or from `https://niblet.pymodel.com/media/`, which 
  • Network accessline 111
    `https://niblet-api.pymodel.com/mcp`
  • Network accessline 115
    [Credential-related source omitted]

skill/niblet/references/evidence.md

  • Network accessline 9
    Niblet's catalogue site at [niblet.pymodel.com](https://niblet.pymodel.com) is a human browsing surface for signed-in accounts. Agent retrieval goes through the tools and the keyed REST fallback in [the connection guide](connection.md). Use

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Signal confidence: 19 facts

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeCompatibleStandard SKILL.md skill/niblet/SKILL.md.claude/skills/<skill-name>/
CodexNativeManifest AGENTS.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md skill/niblet/SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md skill/niblet/SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md skill/niblet/SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md skill/niblet/SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.