research-wiki
Persistent research knowledge base that accumulates papers, ideas, experiments, claims, and their relationships across the entire research lifecycle. Inspired by Karpathy's LLM Wiki pattern. Use when user says "知识库", "research wiki", "add paper", "wiki query", "查知识库", or wants to build/query a persistent field map.
85
83%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Quality
Discovery
89%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 excels in trigger term coverage (including bilingual terms) and completeness with an explicit 'Use when' clause. The main weakness is that the specific capabilities could be more concretely enumerated — listing discrete actions like 'add papers', 'query relationships', 'link claims to evidence' would strengthen specificity beyond the current 'accumulates' framing.
Suggestions
Replace the general 'accumulates papers, ideas, experiments, claims, and their relationships' with more concrete action verbs, e.g., 'Adds and indexes research papers, logs experiment results, links claims to supporting evidence, and queries accumulated knowledge.'
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | The description names the domain (research knowledge base) and mentions what it accumulates (papers, ideas, experiments, claims, relationships), but the concrete actions are somewhat vague — 'accumulates' and 'relationships' are not as specific as listing discrete operations like 'add paper', 'query entries', 'link claims to experiments'. | 2 / 3 |
Completeness | Clearly answers both 'what' (persistent research knowledge base that accumulates papers, ideas, experiments, claims, and relationships across the research lifecycle) and 'when' (explicit 'Use when' clause with specific trigger phrases and a general condition about building/querying a persistent field map). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including both English and Chinese variations: '知识库', 'research wiki', 'add paper', 'wiki query', '查知识库', 'persistent field map'. These are terms users would naturally say when needing this skill. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche — a persistent research wiki/knowledge base inspired by Karpathy's LLM Wiki pattern. The combination of bilingual trigger terms, the specific 'research lifecycle' framing, and the named pattern make it unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 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 comprehensive, highly actionable skill that provides concrete CLI commands, exact schemas, and well-sequenced workflows with validation checkpoints. Its main weakness is verbosity — at 300+ lines with inline integration hooks and detailed pseudocode for three separate hook scenarios, it could benefit from splitting advanced content into referenced files. The content is well-structured with tables and clear sections, but the sheer volume taxes the token budget.
Suggestions
Extract the three integration hooks (Hook 1-3) into a separate file like `integration-hooks.md` and reference it from the main skill, reducing inline bulk by ~40%.
Remove the duplicate Karpathy acknowledgement at the bottom — it's already mentioned in the Overview section.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is quite long (~300+ lines) with some redundancy — the Karpathy reference appears twice, integration hooks are verbose pseudocode that could be tightened, and the overview/core concepts sections explain things Claude could infer. However, the tables and schemas are dense and earn their tokens. | 2 / 3 |
Actionability | Provides fully executable CLI commands for every subcommand (ingest, sync, add_edge, lint), complete paper page schema with exact frontmatter fields, concrete examples with real-looking arXiv IDs, and specific integration hook pseudocode that is detailed enough to implement directly. | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced with explicit validation — lint checks for orphans/contradictions/stale claims, the ingest flow has dedup checks, query_pack has deterministic pruning priority, and integration hooks specify exact ordering (e.g., 'After ideation — THIS IS CRITICAL'). The re-ideation trigger section provides clear thresholds for when to act. | 3 / 3 |
Progressive Disclosure | References shared-references/integration-contract.md appropriately, but the skill itself is monolithic — the lengthy integration hooks, full page schema, and detailed query_pack budget table could be split into separate reference files. For a skill this complex, more content should be offloaded to supporting documents rather than inlined. | 2 / 3 |
Total | 10 / 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 |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
2028ac4
- 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.