cyberneticlibrary

The library

Everything we index — ranked by what works, never by stars.

WORKS 48
Record research decisions and experimentsskillDataOpsL2
ara-research-manager · Manage multi-version research artifact evolution with checkpoints, exploration branching, and job resumption tracking.
WORKS 48
Review research artifact rigorskillProductDataL2
ara-rigor-reviewer · Verify ARA claims are grounded in actual source evidence, detect hallucinations, and flag unsourced assertions before publication.
WORKS 48
Analyze chaotic systems and emergenceskillDataL1
mfe-emergence · Analyzing chaos, bifurcations, fractals, and emergent behavior in nonlinear dynamical systems with quantified complexity metrics.
WORKS 48
Analyze periodic phenomenaskillDataL1
mfe-waves · Analyzing complex functions and signals using Fourier theory, residue calculus, and analytic continuation.
WORKS 48
Analyze AlphaFold protein structuresskillDataL2
572-usage-guide_b6c0a8e7 · Fetches AI-predicted protein structures with B-factor confidence scores for structural analysis.
WORKS 48
Apply Gaussian regularization to embeddingsskillDataL2
sigreg · When sigreg guidance needs expert judgment beyond generic alternatives.
WORKS 53
Run deep research with citationsskillDataL2
perplexity-2 · Getting AI-synthesized answers grounded in current web sources with citations, faster than manual research.
WORKS 48
Optimize BigQuery compute slot costsskillFinanceDataL3
optimize-model-compute · Balancing predictable SLA-critical workloads (reserved) with cost-flexible ad-hoc queries (on-demand).
WORKS 48
Analyze detection coverage rulesskillOpsDataL3
detection-coverage-analysis · Quickly finding detection gaps in ransomware/APT/persistence tactics without token waste.
WORKS 48
Measure and optimize agent growthskillDataProductL3
agent-analytics · Running lightweight A/B tests on headlines and CTAs without backend infrastructure changes when you want agents to own experiment decisions
WORKS 48
Process PDFs at production scaleskillDataOpsL3
PDF Processing Pro · Batch processing structured PDFs (forms, reports) in production when you need robust error handling and type validation
WORKS 48
Automate Amplitude event trackingskillDataProductL2
amplitude-automation · Programmatically sending custom events to Amplitude and managing user cohorts when you need machine-readable user property operations
WORKS 48
Query crypto market dataskillFinanceDataL2
coinpaprika-api · Getting current and historical crypto prices without authentication.
WORKS 48
Monitor Datadog metrics and logsskillEngineeringDataL2
datadog-automation · Query production metrics and logs with alert creation across multi-service infrastructure.
WORKS 48
Query DeFi token and pool dataskillFinanceDataL2
dexpaprika-api · Query cross-chain DeFi liquidity and token prices across 34+ blockchains in one request.
WORKS 52
Automate Google Analytics reportingskillDataMarketingL2
google-analytics-automation · Export GA4 data for custom analysis, funnel analysis, and pivot reports without logging into Google Analytics UI.
WORKS 48
Automate Google Sheets data operationsskillDataOpsL2
googlesheets-automation · Batch-append data from APIs or CSVs to shared sheets, or query sheets as a lightweight database.
WORKS 48
Automate Mixpanel analytics trackingskillDataProductL2
mixpanel-automation · Real-time analytics dashboards and behavior segmentation for multi-channel product validation.
WORKS 52
Track AI news across 80+ sourcesskillDataProductL2
morning-ai · AI researchers and founders who need daily curated AI news across 80+ entities without managing multiple newsletters.
WORKS 48
Automate PostHog analytics eventsskillDataL2
posthog-automation · Automating multi-step SaaS workflows via unified MCP toolkit with consistent error handling.
WORKS 48
Automate Segment user trackingskillDataL2
segment-automation · Automating multi-step SaaS workflows via unified MCP toolkit with consistent error handling.
WORKS 48
Track trending topics over rolling windowsskillDataL2
last30days — Rolling Topic Research Skill · When answering 'what happened with X this month' beats a single web search by merging high-signal sources with dated context
WORKS 48
Update research citation countsskillDataL1
update-cite-count · When maintaining rankings of research papers that should remain sorted by current citation count without manual data entry
WORKS 48
Detect fake reviews on WalmartskillProductDataL1
walmart-review-checker · When validating Walmart product review integrity or detecting review manipulation.
WORKS 48
Run genomics sequencing workflowsskillDataL3
ngs-epigenomics-peaks · Dispatching ATAC-seq, ChIP-seq, CUT&RUN, and CUT&Tag analysis to assay-specific QC, alignment, peak-calling, and differential workflows.
WORKS 48
Analyze scenarios and predict future outcomesskillDataL1
hard-predict-future · Future-oriented questions requiring quantitative rigor
WORKS 48
Model future possibilities and scenariosskillDataL1
soft-predict-future · Prediction questions that need scenario planning
WORKS 48
Process and export Excel spreadsheetsskillDataL2
document-processing-xlsx · Spreadsheet automation and data extraction
WORKS 48
Enrich data via web searchskillDataL2
web_search_enricher · Filling postal code → municipality gaps in data records when deterministic database lookups fail and confidence is low.
WORKS 48
Analyze financial text sentimentskillDataFinanceL2
alphaear-sentiment · Scoring sentiment on bulk financial news when local FinBERT speed outweighs LLM latency.
WORKS 52
Access Google Search Console and AnalyticsskillMarketingDataL2
seo-data · Querying real SEO metrics (GSC/GA4/Bing) directly against user's own verified properties.
WORKS 48
Extract and analyze Bilibili commentsskillDataL2
bbc-skill · Quantifying audience sentiment when qualitative comment analysis beats binary metrics.
WORKS 48
Extract clinical NLP entitiesskillDataL2
clinical-nlp-extractor · Extracting structured medical data from unstructured clinical notes at scale.
WORKS 48
Visualize evaluation progress over cyclesskillDataL2
os-improvement-report · Tracking and visualizing agent loop improvement trends over cycles
WORKS 48
Audit vector database coverage gapsskillEngineeringDataL2
vector-db-audit · When you need O(1) keyword lookup across RLM summaries without semantic search overhead.
WORKS 48
Clean up orphaned vector database chunksskillEngineeringDataL2
vector-db-cleanup · When RLM Phase 1 keyword search is insufficient and semantic nearest-neighbor matching is needed.
WORKS 48
Index files into vector databaseskillEngineeringDataL2
vector-db-ingest · When RLM Phase 1/2 results are insufficient and you need precise regex or exact-match lookup.
WORKS 48
Build and backtest trading modelsskillFinanceDataL3
quant-analyst · When backtesting trading strategies with proper risk metrics and out-of-sample validation.
WORKS 48
Analyze KEGG pathways and drug targetsskillDataProductL3
kegg-analysis · When performing multi-step bioinformatics analysis (enrichment, drug targets, cross-species comparison).
WORKS 48
Read and write Google SheetsskillOpsDataL2
sheets-cli · When you need programmatic control of Google Sheets without leaving the terminal.
WORKS 48
Render research chartsskillDataMarketingL1
Chart Visualization · When you need to communicate data insights through visual formats.
WORKS 48
Download models from HuggingFaceskillEngineeringDataL2
hf-download · Fetching HuggingFace models with automatic exponential backoff on rate limits.
WORKS 48
Connect to HuggingFace datasetsskillEngineeringDataL2
hf-init · One-time setup of HuggingFace datasets with standard ADR 081 structure.
WORKS 48
Upload snapshots to HuggingFaceskillEngineeringDataL2
hf-upload · Persisting agent learnings and semantic caches to HuggingFace with tags.
WORKS 48
Design batch-aware experimentsskillDataL2
bio-experimental-design-batch-design · Biologists designing multi-batch sequencing to make batch effects correctable
WORKS 48
Convert Jupyter notebooks to marimoskillEngineeringDataL1
jupyter-to-marimo · Converting Jupyter notebooks to marimo interactive Python scripts.
WORKS 48
Evaluate research findings and design experimentsskillProductDataL1
scientific-thinking · Using scientific thinking when use when interpreting research findings, evaluating scientific evidence, analyzing mechanisms, comparing competing hypotheses, designing experiments, or constructing scientific arguments..
WORKS 48
Optimize BigQuery costs and usageskillDataL1
bigquery-cost-optimization · Switching between on-demand and slots pricing with cost-per-query benchmarking.
WORKS 48
Optimize slow BigQuery queriesskillDataL1
bigquery-optimization · Query performance tuning when existing queries exceed slot budget or memory limits.
WORKS 51
Generate BigQuery SQL from textskillDataL1
bigquery-query-generation · Auto-generating idiomatic BigQuery SQL from natural language requirements.
page 8 / 11