create-client-tool
MUST be used whenever creating an AtlasTool (client-side tool) for an Atlas agent. Do NOT manually write AtlasTool definitions or wire them into useAtlasChat — this skill handles the TypeBox schema, execute function, and hook wiring. Prerequisite: integrate-atlas-chat (vendored src/atlas-agent + TypeBox/AJV deps). This includes tools that fetch data, render UI, call APIs, show charts, query local state, or perform any browser-side action. Triggers: AtlasTool, client tool, add tool, create tool, new tool, tool definition, agent tool.
67
82%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong, well-crafted skill description that clearly defines its scope, provides explicit trigger terms, and establishes both what it does and when it should be used. The inclusion of prerequisites, concrete actions, and a dedicated triggers list makes it highly effective for skill selection. The imperative 'MUST be used' and 'Do NOT manually write' framing adds clarity about exclusivity.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: creating AtlasTool definitions, handling TypeBox schema, execute function, hook wiring. Also enumerates concrete use cases like fetching data, rendering UI, calling APIs, showing charts, querying local state. | 3 / 3 |
Completeness | Clearly answers both 'what' (creates AtlasTool definitions with TypeBox schema, execute function, and hook wiring) and 'when' (explicit 'MUST be used whenever creating an AtlasTool' plus a triggers list). Also includes prerequisite information and scope of applicability. | 3 / 3 |
Trigger Term Quality | Includes a dedicated 'Triggers:' section with natural terms users would say: 'AtlasTool', 'client tool', 'add tool', 'create tool', 'new tool', 'tool definition', 'agent tool'. These cover common variations well. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — targets a very specific niche (AtlasTool client-side tools for Atlas agents) with domain-specific terminology like TypeBox, AJV, useAtlasChat, and Atlas agent. Unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, actionable skill with executable code examples and a clear step-by-step structure for creating Atlas client tools. Its main weaknesses are the lack of validation/verification checkpoints (no 'test it works' step) and some unnecessary background explanation that Claude doesn't need. The content is well-organized but could benefit from a verification step and slight trimming.
Suggestions
Add a verification step after wiring (e.g., 'Step 5 — Verify: run the dev server, open the chat, and confirm the agent can invoke the tool without schema errors') to close the feedback loop.
Remove or significantly trim the 'Background' section — Claude understands client-tool execution flows and doesn't need the 5-step conceptual explanation.
Add a brief troubleshooting note for common failure modes (e.g., incorrect import paths, schema validation errors at runtime) to improve error recovery.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary explanation (e.g., the 'Background' section explaining the client tool flow is largely conceptual knowledge Claude already has). The TypeBox quick reference table is useful but borderline — Claude knows TypeBox. The core steps are reasonably tight but could be trimmed. | 2 / 3 |
Actionability | Provides fully executable, copy-paste-ready TypeScript code for defining a tool, wiring it into useAtlasChat, and rendering tool results. The code examples are complete with proper imports, types, and realistic patterns. The TypeBox reference table adds concrete guidance. | 3 / 3 |
Workflow Clarity | The four steps are clearly sequenced and logical (understand → define → wire → render). However, there are no validation or verification checkpoints — no step to test that the tool actually works, no feedback loop for common errors (e.g., schema validation failures, import path issues), and no explicit 'verify it compiles' step. | 2 / 3 |
Progressive Disclosure | The content is well-structured with clear sections and headers, but everything is inline in a single file with no references to supporting documents. The TypeBox reference table and the rendering section could be split out. For a skill of this length (~100 lines), it's acceptable but not optimal — no bundle files are provided to offload detail. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
metadata_version | 'metadata.version' is missing | Warning |
Total | 9 / 11 Passed | |
- Repository
- cognitedata/builder-skills
- Commit
d6af887
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.