Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Research Knowledge Acquisition Engine with 5 campaigns (literature-survey, patent-mining, benchmark-archaeology, meta-analysis, baseline-establishment). Use this skill whenever a user needs to systematically acquire research knowledge — academic literature, patent landscapes, benchmark evaluations, cross-study statistical synthesis, or SOTA performance baselines. Pre-condition: north-star-crystallization must be complete.
.claude/skills/yogsoth-ai-knowledge-acquisition/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✗→✓ | ▲ Improved | 2% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -2% | 0% |
| case-07 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-08 | ✗→✓ | ▲ Improved | -3% | 0% |
| case-10 | ✗→✓ | ▲ Improved | -40% | 0% |
Systematic research knowledge acquisition engine. Five campaigns, each a self-contained autonomous research activity domain. You provide a research intent — the engine routes to the right campaign, selects a strategy, and executes autonomously with quantitative budget enforcement.
North-star-crystallization must be complete before entering any campaign. Research intent must be fully crystallized.
ENTRY.md (this file)
→ Campaign (5): self-contained research activity domain
→ Strategy: selected by analysis purpose/intent
→ Tactic: multi-step orchestration pattern (reusable across strategies)
→ SOP: single operation (import or subagent)| Signal | Campaign | |--------|----------| | literature review, survey, paper search, PRISMA, snowball | → literature-survey | | patent analysis, prior art, white space, claims, IPC | → patent-mining | | benchmark analysis, evaluation methods, metric flaws, leaderboards, saturation | → benchmark-archaeology | | cross-study statistical synthesis, effect size, heterogeneity, publication bias, GRADE | → meta-analysis | | SOTA compilation, performance comparison, baseline reproduction, progress curves | → baseline-establishment |
Campaigns can be composed:
The orchestrator decides composition based on the crystallized North Star statement.
| MCP Server | Tools | |------------|-------| | brave-search | brave_web_search, brave_news_search, brave_llm_context | | apify | rag-web-browser, google-scholar-scraper | | alphaxiv | discover_papers, get_paper_content, answer_pdf_queries, read_files_from_github_repository | | semantic-scholar | ss_paper, ss_paper_batch, ss_references, ss_citations, ss_recommendations, ss_relevance_search, ss_author, ss_author_papers |
| Dependency | What It Provides | |-----------|-----------------| | web-browsing | web-search + web-research | | literature-engine | literature-overview + literature-search + literature-research | | subagent-spawning | Subagent dispatch conventions | | context-management | Checkpoint protocol |
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Campaign | When to use | | --- | --- | | baseline-establishment | SOTA Performance Baseline Campaign — 5 strategies for systematically collecting, standardizing, and analyzing performance data across methods. Produces standardized comparison tables, progress curves, and headroom analysis. | | benchmark-archaeology | Evaluation Methodology Archaeology Campaign — 5 strategies for systematic analysis of AI/ML benchmarks, metrics, and leaderboards. Reveals construct validity issues, saturation, data contamination, and evaluation protocol inconsistencies. | | literature-survey | Autonomous Literature Survey Campaign — 5 research paradigms (scoping, systematic, deep, narrative, snowball) with quantitative budget enforcement. Selects and executes the right survey paradigm based on research intent. | | meta-analysis | Cross-Study Statistical Synthesis Campaign — 5 strategies for systematic collection and methodological planning of multi-study evidence synthesis. Covers pairwise, network, cumulative meta-analysis, heterogeneity investigation, and bias detection. Stops at protocol design (no computation). | | patent-mining | Systematic Patent Analysis Campaign — 5 strategies for patent landscape analysis, prior art search, white space identification, competitive intelligence, and claim analysis. Produces structured patent intelligence reports. |
<!-- END available-tables (generated) -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 44,073 | 39,356 | -11% | 1 | 1 | 0% | 7,235 | 6,420 | -11% | 0 | 0 | — |
case-02 | fail→fail | 60,813 | 35,709 | -41% | 1 | 1 | 0% | 8,242 | 5,653 | -31% | 0 | 0 | — |
case-03 | fail→fail | 51,692 | 36,230 | -30% | 1 | 1 | 0% | 8,265 | 6,874 | -17% | 0 | 0 | — |
case-04 | fail→fail | 21,565 | 20,993 | -3% | 1 | 1 | 0% | 2,935 | 3,444 | +17% | 0 | 0 | — |
case-05 | fail→pass | 20,982 | 22,435 | +7% | 1 | 1 | 0% | 2,482 | 2,532 | +2% | 0 | 0 | — |
case-06 | fail→pass | 18,614 | 18,692 | +0% | 1 | 1 | 0% | 3,243 | 3,168 | -2% | 0 | 0 | — |
case-07 | fail→pass | 17,360 | 5,587 | -68% | 1 | 1 | 0% | 2,143 | 1,836 | -14% | 0 | 0 | — |
case-08 | fail→pass | 20,335 | 12,915 | -36% | 1 | 1 | 0% | 2,449 | 2,373 | -3% | 0 | 0 | — |
case-09 | pass→pass | 18,456 | 8,740 | -53% | 1 | 1 | 0% | 3,182 | 2,459 | -23% | 0 | 0 | — |
case-10 | fail→pass | 21,498 | 2,632 | -88% | 1 | 1 | 0% | 2,404 | 1,431 | -40% | 0 | 0 | — |
case-11 | fail→pass | 13,030 | 8,509 | -35% | 1 | 1 | 0% | 2,152 | 1,574 | -27% | 0 | 0 | — |
case-12 | fail→fail | 21,180 | 8,055 | -62% | 1 | 1 | 0% | 2,816 | 1,489 | -47% | 0 | 0 | — |
case-13 | pass→pass | 22,357 | 8,402 | -62% | 1 | 1 | 0% | 2,935 | 1,527 | -48% | 0 | 0 | — |
case-14 | fail→pass | 16,026 | 8,411 | -48% | 1 | 1 | 0% | 1,786 | 1,525 | -15% | 0 | 0 | — |
case-15 | fail→pass | 24,272 | 9,719 | -60% | 1 | 1 | 0% | 2,354 | 1,781 | -24% | 0 | 0 | — |
case-16 | pass→pass | 14,932 | 8,411 | -44% | 1 | 1 | 0% | 1,606 | 1,603 | -0% | 0 | 0 | — |
case-17 | pass→pass | 20,211 | 7,066 | -65% | 1 | 1 | 0% | 2,341 | 1,287 | -45% | 0 | 0 | — |
case-18 | fail→pass | 45,915 | 6,837 | -85% | 1 | 1 | 0% | 3,479 | 1,243 | -64% | 0 | 0 | — |
case-19 | fail→pass | 6,756 | 8,150 | +21% | 1 | 1 | 0% | 1,052 | 1,434 | +36% | 0 | 0 | — |
case-20 | fail→pass | 38,269 | 2,073 | -95% | 1 | 1 | 0% | 1,524 | 1,319 | -13% | 0 | 0 | — |
case-21 | pass→pass | 21,191 | 7,997 | -62% | 1 | 1 | 0% | 2,747 | 1,451 | -47% | 0 | 0 | — |
case-22 | pass→fail | 25,133 | 50,871 | +102% | 1 | 1 | 0% | 4,516 | 9,217 | +104% | 0 | 0 | — |
case-23 | pass→pass | 17,617 | 26,554 | +51% | 1 | 1 | 0% | 2,132 | 4,617 | +117% | 0 | 0 | — |
case-24 | pass→pass | 32,948 | 20,728 | -37% | 1 | 1 | 0% | 2,112 | 3,381 | +60% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 24 cases were attempted, and 23 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +42 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.