dm-limits-and-best-practices
Reference skill for CDF Data Modeling API best practices. Covers concurrency limits (avoiding 429s), pagination patterns for instances.list and instances.query, batching write operations, search vs filter guidance, and the QueuedTaskRunner (Semaphore) utility for controlling concurrent requests. Triggers: DMS limits, 429 error, rate limit, pagination, cursor, nextCursor, batching, semaphore, QueuedTaskRunner, cdfTaskRunner, instances.search, instances.list, instances.query, instances.upsert, concurrency, deadlock.
89
88%
Does it follow best practices?
Impact
96%
1.47xAverage score across 3 eval scenarios
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 skill description that clearly defines its scope around CDF Data Modeling API best practices with specific, actionable topics. The explicit 'Triggers:' section provides comprehensive keyword coverage including error codes, API methods, and utility names. The description is well-structured, concise, and highly distinguishable from other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and concepts: concurrency limits (avoiding 429s), pagination patterns for instances.list and instances.query, batching write operations, search vs filter guidance, and the QueuedTaskRunner (Semaphore) utility for controlling concurrent requests. | 3 / 3 |
Completeness | Clearly answers 'what' (CDF Data Modeling API best practices covering concurrency, pagination, batching, search vs filter, QueuedTaskRunner) and 'when' via an explicit 'Triggers:' clause listing specific terms that should activate this skill. | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including error codes (429), API method names (instances.list, instances.query, instances.upsert, instances.search), utility names (QueuedTaskRunner, cdfTaskRunner, Semaphore), and conceptual terms (rate limit, pagination, cursor, nextCursor, batching, concurrency, deadlock, DMS limits). | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche around CDF Data Modeling API specifics. The trigger terms are domain-specific (QueuedTaskRunner, cdfTaskRunner, instances.search, DMS limits) and unlikely to conflict with generic coding or API skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, highly actionable reference skill with excellent executable examples and clear workflow patterns for CDF Data Modeling APIs. Its main weakness is length — the full semaphore source code and some repeated patterns (pagination shown in multiple sections) inflate the token cost. The content would benefit from splitting the QueuedTaskRunner implementation into a separate file and reducing redundancy between the main sections and the Common Pitfalls section.
Suggestions
Extract the QueuedTaskRunner source code into a separate file (e.g., semaphore.ts) and reference it from SKILL.md rather than inlining ~80 lines of implementation code.
Reduce redundancy between the main Pagination section and Common Pitfalls #2 (Forgetting Pagination) — the same pattern is demonstrated twice with nearly identical code.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient and domain-specific, but it's quite long (~350 lines). The full QueuedTaskRunner source code (~80 lines) could be in a separate file. Some sections like the decision guide table and the search examples are well-targeted, but the overall volume is high for a reference skill. The 'Common Pitfalls' section repeats patterns already shown above (e.g., pagination is demonstrated twice). | 2 / 3 |
Actionability | Excellent actionability throughout. Every pattern includes fully executable TypeScript code examples that are copy-paste ready. The chunking utility, pagination patterns, batched upsert/delete, semaphore usage, and search vs filter examples are all concrete and complete with realistic API call structures. | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced with explicit validation patterns. The pagination workflows show clear loop structures with cursor checks, the batching workflow shows chunk→schedule→await, and the Common Pitfalls section provides explicit BAD/GOOD comparisons that serve as validation guidance. The summary checklist acts as a verification checkpoint. Deadlock prevention is explicitly addressed with clear examples. | 3 / 3 |
Progressive Disclosure | The content is well-organized with clear section headers and a logical flow, but it's monolithic — the full QueuedTaskRunner source code (~80 lines) should be in a separate referenced file rather than inline. There are no bundle files to offload content to, and the skill doesn't reference any supporting files. For a skill this long, splitting the semaphore implementation and detailed examples into separate files would improve token efficiency. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (614 lines); consider splitting into references/ and linking | Warning |
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
metadata_version | 'metadata.version' is missing | Warning |
Total | 8 / 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.