alterlab-skill-finder

$npx mdskill add AlterLab-IEU/AlterLab-Academic-Skills/alterlab-skill-finder

Routes generic AlterLab requests to the correct skill or launches multi-agent workflows.

  • Handles vague or multi-stage AlterLab requests when no specific skill is named.
  • Reads the bundled skill index and uses host subagent/Workflow tools for orchestration.
  • Asks clarifying questions to select skills for multi-step goals or alterflow keywords.
  • Returns a routed task or executes a dynamic multi-agent workflow composing skills.

SKILL.md

.github/skills/alterlab-skill-finderView on GitHub ↗
---
name: alterlab-skill-finder
description: "The AlterLab front door and multi-agent launcher — routes a task to the right AlterLab skill(s) when the user invokes the suite without naming one, and for a multi-stage goal (or on the keyword 'alterflow', aliases 'alterresearch' / 'ultralab') it CLARIFIES the goal with a few questions, SELECTS the skills the task needs, and runs a dynamic multi-agent workflow composing them (via alterlab-workflow-orchestration, alterlab-research-pipeline, or alterlab-ssci-orchestrator). Triggers on 'use AlterLab skills', 'which AlterLab skill for X', 'is there an AlterLab skill for…', a multi-stage research goal, 'alterflow …', or any generic AlterLab request where the user does not know skill names. It always asks clarifying questions before executing a multi-step run. Use when someone references AlterLab generically, describes a multi-stage goal, or fires the alterflow keyword; when the user already names a specific skill, defer to that skill directly. Part of the AlterLab Academic Skills suite."
license: MIT
allowed-tools: Read Task
compatibility: "No API key or network required. A routing + orchestration front-end: it reads the bundled skill index and hands off to the matching AlterLab skill(s). Multi-agent execution uses the host's subagent/Workflow tools where available (Claude Code, Cowork); on surfaces without them it decomposes the work into sequential phases."
metadata:
    skill-author: AlterLab
    version: "1.0.0"
    depends_on: "dispatches to any AlterLab skill; composes multi-step runs via alterlab-research-pipeline, alterlab-workflow-orchestration, and alterlab-ssci-orchestrator"
---

# AlterLab Skill Finder — Name the Task, Get the Right Skill (or the Whole Workflow)

**Skill type: ROUTER / LAUNCHER.** Users say *"use AlterLab skills"* or fire **`alterflow`** without
knowing the 230+ skill names. This is the front door: it reads the task, picks the AlterLab skill(s)
that fit, and either **uses one skill** or **launches a clarified, multi-agent workflow**.

## Core Mission

```
WHEN THE USER NAMES ALTERLAB BUT NOT A SKILL, YOU PICK THE SKILL(S).
SIMPLE TASK → ROUTE TO ONE SKILL.   BIG / KEYWORD TASK → CLARIFY, THEN ORCHESTRATE MANY.
ALWAYS ASK YOUR QUESTIONS BEFORE YOU START A MULTI-STEP RUN.
```

## Two modes

| Mode | Fires on | What happens |
|------|----------|--------------|
| **Route** (default) | "use AlterLab skills to…", "which AlterLab skill for…", any generic AlterLab ask | classify the task → map to a domain → name and **apply** the best-fitting skill(s) — Anthropic's **routing** pattern |
| **Orchestrate** | a task that **spans several dependent stages** (research → write → review, etc.); the keyword **`alterflow`** — aliases `alterresearch` / `ultralab` — is an explicit shortcut into this mode | **clarify → select skills → plan a multi-agent workflow → confirm → execute** — Anthropic's **orchestrator-workers** pattern |

## When to Use This Skill

- "Use AlterLab skills to [do X]." / "Which AlterLab skill should I use for [task]?"
- "Is there an AlterLab skill / workflow for [task]?"
- "**alterflow** — investigate [topic] and draft a paper." (→ Orchestrate mode)
- Any request that references AlterLab by name without specifying a skill.

### Does NOT Trigger

| The request is really about… | Route to | Why not this skill |
|---|---|---|
| A **specific named** skill ("run alterlab-deep-research", "use survey-analysis") | that skill directly | The user already knows it; no routing needed. |
| Composing a bespoke multi-agent workflow the user is hand-designing | `alterlab-workflow-orchestration` | This skill *calls* that engine; it doesn't replace its mechanics. |
| A task with no AlterLab framing at all | (answer normally) | Nothing to route. |

## Route mode — the method

1. **Classify the task** — what is the user doing (find literature? clean data? fit a model? write a
   section? query a database? design a study?).
2. **Map to a domain** (table below).
3. **Select the skill(s)** — name the best fit. For an exact niche name, consult
   `references/skill_index.md` (every AlterLab skill, grouped by domain, one-liner each) or match the
   installed AlterLab skill descriptions.
4. **Apply it** — invoke the skill and do the work. Tell the user which skill you picked and why (one line).
5. **If the routing genuinely forks, ask ONE clarifying question** before committing.

## Orchestrate mode — `alterflow` (clarify FIRST, then multi-agent)

Fires when the task **spans several dependent stages** (e.g. "investigate X *and* write it up",
"research → analyze → publish") — the `alterflow` keyword is just an explicit shortcut into this
mode, not the only way in; detect the multi-stage intent even when the user never types it. This is
Anthropic's **orchestrator-workers** pattern (a lead agent dynamically decomposes the task, delegates
to workers, and synthesizes) with a human-in-the-loop clarify/confirm gate in front. Do **not** start
executing — run this sequence:

1. **CLARIFY (mandatory, before any execution).** Ask 2–4 sharp scoping questions — e.g. the research
   question / deliverable, quantitative vs qualitative, the dataset or corpus, the target output
   (report, paper, model, figures), depth/time budget, and language. Do not guess when the answer
   changes the plan.
2. **SELECT.** From the answers, choose the AlterLab skills the job needs (use the domain map + the
   index). List them. **First ask whether it should be multi-agent at all:** if the work is tightly
   coupled, needs one shared context, or has little that runs in parallel (Anthropic's caution — most
   coding-like, dependency-heavy tasks), prefer a **sequential pipeline (prompt-chaining)** or a single
   Route-mode skill over a fan-out.
3. **PLAN.** Lay out the **dynamic multi-agent workflow** — the phases, which skill runs in each, where
   subagents fan out in parallel, and the **evaluator-optimizer** verification/critique step. **Scale
   the subagent count to complexity and default low** (Anthropic's rule of thumb: simple fact-finding
   ≈ 1 agent; a focused comparison ≈ 2–4; only genuinely broad work ≈ 10+). Reuse the existing
   orchestrators rather than reinventing them:
   - end-to-end research→publish → **`alterlab-research-pipeline`** (deep-research → paper-writer → paper-reviewer, revision loops)
   - a whole social-science study → **`alterlab-ssci-orchestrator`** (design → measurement/reflexivity → sampling → analysis module → inference)
   - a bespoke fan-out / judge-panel / adversarial-verify workflow → **`alterlab-workflow-orchestration`**
4. **CONFIRM — plan *and its rough cost*.** Show the selected skills + the phases in a few lines,
   **plus an effort estimate** (≈ how many phases / subagents). Multi-agent runs spend far more tokens
   than a single pass, so let the user opt in knowingly; get a go-ahead (or incorporate a correction).
5. **EXECUTE.** Run it as a multi-agent workflow — spawn subagents via the host's Task/Workflow tools
   where available (Claude Code, Cowork); on surfaces without them, execute the phases sequentially and
   keep the same hand-offs. **Give every spawned subagent a complete task spec** — an objective, an
   output format, which skills/tools/sources to use, and clear boundaries — or workers duplicate work
   and leave gaps (`alterlab-workflow-orchestration` models these specs). Carry each stage's artifact
   to the next (the pipelines define the hand-off contracts).

The one rule: **questions before execution.** A short clarify step beats a wrong 20-agent run.

## Domain routing map (17 domains)

| If the task is about… | Domain | Representative skills |
|---|---|---|
| Find literature, fact-check, discover, manage references | **research-tools** / **core** | `alterlab-deep-research`, `alterlab-research-lookup`, `alterlab-pyzotero`, `alterlab-citation-verifier` |
| Write / draft / revise a paper, abstract, grant, poster | **core** / **writing-tools** | `alterlab-paper-writer`, `alterlab-scientific-writing`, `alterlab-grant-writer` |
| Review / critique a manuscript | **core** | `alterlab-paper-reviewer` |
| Query a scientific database (PubMed, ChEMBL, UniProt, GEO, …) | **databases** | `alterlab-pubmed`, `alterlab-chembl`, `alterlab-uniprot`, … (39) |
| Genomics, proteomics, single-cell, structure prediction | **bioinformatics** | `alterlab-scanpy`, `alterlab-alphafold`, `alterlab-biopython`, … |
| Chemistry, drug discovery, docking, ADMET | **cheminformatics** | `alterlab-rdkit`, `alterlab-deepchem`, … |
| Clinical decision support, trials, medical imaging | **clinical-research** | `alterlab-clinicaltrials`, `alterlab-clinical-decision`, … |
| ML, statistics, data analysis, dataframes | **data-science** | `alterlab-scikit-learn`, `alterlab-statistical-analysis`, `alterlab-statsmodels`, `alterlab-pymc` |
| Plots, charts, figures, schematics | **visualization** | `alterlab-matplotlib`, `alterlab-scientific-viz`, `alterlab-infographics` |
| Lab platforms (Benchling, DNAnexus, Opentrons) | **lab-integrations** | `alterlab-benchling`, … |
| Quantum, geospatial, materials, astronomy, digital humanities | **domain-specific** | `alterlab-qiskit`, `alterlab-geopandas`, `alterlab-pymatgen`, … |
| Convert / handle documents, Markdown, notebooks, PDFs | **document-tools** | `alterlab-markitdown`, `alterlab-pdf-explore` |
| Finance, economics, market/financial data | **finance-economics** | `alterlab-fred`, `alterlab-sec-edgar`, … |
| Turkish academic system (YÖK, ÜAK, DergiPark, TÜBİTAK, doçentlik) | **turkish-academia** | `alterlab-dergipark`, `alterlab-tubitak-proposal`, `alterlab-docentlik-eligibility`, … |
| Teaching, IRB, grant admin, accreditation, recommendation letters | **faculty-life** | `alterlab-syllabus-ai-policy`, `alterlab-irb-consent`, … |
| Research-rigor gates (pre-registration, test choice, transparency) | **methodology** | `alterlab-test-selection-guard`, `alterlab-preregistration-discipline` |
| Design/run a whole **social-science study** (survey, qualitative, causal, multilevel, meta, missing data) | **social-science-workflow** | `alterlab-ssci-orchestrator` (+ 16 gates & modules) |

Full per-domain listing of every skill with a one-liner: `references/skill_index.md`.

## Output pattern

```
Task: <one-line restatement>
Mode: route | orchestrate
AlterLab skill(s): <name(s)> — <why this fits>
[orchestrate] Questions first: <2–4 scoping questions>   ← ask, then wait
[orchestrate] Plan: <phase 1 skill → phase 2 skills (fan-out) → verify → deliver>
→ on confirmation, execute.
```

## References

- `references/skill_index.md` — every AlterLab skill, grouped by domain, one-liner each (generated, kept in sync with the catalog).

Part of the AlterLab Academic Skills suite.

More from AlterLab-IEU/AlterLab-Academic-Skills

SkillDescription
alterlab-abm-mesaBuilds agent-based models of social systems with Mesa 3 — the current AgentSet API (model.agents.shuffle_do('step'), auto-assigned unique_id, mandatory super().__init__(seed=...)), cell spaces (mesa.discrete_space OrthogonalMooreGrid / classic mesa.space grids), the DataCollector, batch_run parameter sweeps, and SolaraViz — for emergence, segregation, diffusion, opinion dynamics, and cooperation models. It uses the Mesa 3.x API (the old mesa.time schedulers like RandomActivation are removed) and treats the model as a generative theory to be validated, not just run. Use when the request mentions an agent-based model, Mesa, simulating interacting agents, or emergent macro behavior from micro rules. For discrete-event (queueing/process) simulation prefer alterlab-simpy; for reinforcement learning prefer alterlab-stable-baselines3. Part of the AlterLab Academic Skills suite.
alterlab-academic-careerPrepares academic career documents and professional-development materials for faculty and researchers. Use when drafting academic CVs, research statements, teaching philosophies, diversity statements, faculty-position cover letters, tenure dossiers, promotion narratives, academic portfolios, or mentorship statements, and when planning conference-networking strategy, building academic web presence, setting up ORCID profiles, or understanding impact metrics. Part of the AlterLab Academic Skills suite.
alterlab-accreditation-aolScaffolds program-level Assurance-of-Learning (AoL) documentation for AACSB (2020 Standard 5) and ABET (Criterion 3 Student Outcomes, Criterion 4 Continuous Improvement) accreditation — program learning outcomes / competency goals, curriculum-to-outcome mapping matrices, direct- and indirect-assessment plans, rubric design, and closing-the-loop continuous-improvement narratives — and validates the structure of an outcome-mapping matrix with scripts/aol_matrix.py. Use when the user needs AACSB or ABET assurance-of-learning material, a program-learning-outcomes set, a curriculum/outcome map or coverage matrix, a direct/indirect assessment plan, a closing-the-loop report, or accreditation self-study text. For single-course design or course rubrics prefer alterlab-teaching-design; for post-award grant reports prefer alterlab-grant-reporting. Part of the AlterLab Academic Skills suite.
alterlab-adaptyvSubmits and tracks protein-testing experiments on the Adaptyv Bio Foundry cloud lab (wet-lab validation), and optimizes protein sequences before submission with computational tools (NetSolP, SoluProt, SolubleMPNN, ESM). Use when designing proteins that need wet-lab validation - binding/affinity screening, expression testing, thermostability, or fluorescence assays - or when submitting experiments to the Foundry API, browsing the target catalog, tracking experiment status, retrieving results, or pre-screening sequences for solubility/expression. Triggers on "Adaptyv", "Foundry API", "cloud lab", "biolayer interferometry / BLI", "wet-lab validation". Part of the AlterLab Academic Skills suite.
alterlab-aeonRuns time series machine learning with the aeon library — classification, regression, clustering, forecasting, anomaly detection, segmentation, and similarity search via scikit-learn compatible APIs. Use when working with temporal data, sequential patterns, or time-indexed observations (univariate or multivariate) that need specialized algorithms beyond standard ML approaches. Part of the AlterLab Academic Skills suite.
alterlab-akademik-tesvikComputes the annual Turkish academic-incentive (akademik teşvik) score for Devlet (state) university öğretim elemanı (academic staff) from a classified activity list, applying the verified Akademik Teşvik Ödeneği Yönetmeliği (2018/11834) Faaliyet Hesaplama Tablosu headline puanları (Proje 20, Araştırma 15, Yayın 30, Patent 30, Atıf 30, Tebliğ 20, Ödül 20) with the k (author-count), p (Q1 1/Q2 0.8/Q3 0.5/Q4 0.25), and r (project-role) coefficients via türü puanı = Σ(faaliyet oranları) × headline (MADDE 8/2), then enforcing the per-type headline ceilings, the 100 cap (MADDE 8/3), and the net-30 payment gate (MADDE 10/3) in a deterministic scripts/tesvik_score.py. Use when the user wants to calculate akademik teşvik puanı, check the en az 30 (net-30) threshold, apply TABLO 4 ceilings, or estimate teşvik eligibility; keeps its coefficients separate from doçentlik scoring. For doçentlik point eligibility prefer alterlab-docentlik-eligibility. Part of the AlterLab Academic Skills suite.
alterlab-alpha-vantageAccesses real-time and historical stock market data, forex rates, cryptocurrency prices, commodities, economic indicators, and 50+ technical indicators via the Alpha Vantage API (requires a free API key from alphavantage.co). Use when fetching stock prices (OHLCV), company fundamentals (income statement, balance sheet, cash flow), earnings, options data, market news/sentiment, insider transactions, GDP, CPI, treasury yields, gold/silver/oil prices, Bitcoin/crypto prices, forex exchange rates, or calculating technical indicators (SMA, EMA, MACD, RSI, Bollinger Bands). Part of the AlterLab Academic Skills suite.
alterlab-alphafoldPredict protein 3D structures with AlphaFold2 via ColabFold — MMseqs2-accelerated MSAs, monomer and AlphaFold2-Multimer complex folding, and confidence-based validation (pLDDT, pTM/ipTM, PAE). Use when folding a protein sequence or complex from FASTA, generating a predicted structure with confidence metrics, ranking models, or checking self-consistency of a design. For co-folding a protein WITH a small-molecule ligand or predicting binding affinity prefer alterlab-boltz; for antibody–antigen or one-FASTA multi-entity complexes prefer alterlab-chai; to LOOK UP an already-computed structure prefer alterlab-alphafold-db; for ESM embeddings or inverse folding prefer alterlab-esm. Part of the AlterLab Academic Skills suite.
alterlab-alphafold-dbAccess the AlphaFold DB of 200M+ AI-PREDICTED protein structures — retrieve models by UniProt accession, download PDB/mmCIF files, and analyze prediction confidence metrics (pLDDT, PAE). Use when a UniProt ID needs a computationally predicted 3D structure or when no experimental structure exists, for homology modeling, protein engineering, or structure-based drug discovery; for EXPERIMENTALLY determined structures (X-ray, cryo-EM, NMR) prefer alterlab-pdb, and for protein sequences, annotations, or accession ID mapping prefer alterlab-uniprot instead. Part of the AlterLab Academic Skills suite.
alterlab-anndataBuild, slice, concatenate, read, and write AnnData annotated data matrices (obs, var, X, layers, obsm, uns) — the scverse data STRUCTURE, not an analysis pipeline. Use when creating or wrangling .h5ad/zarr files, managing cell and gene annotations, concatenating batches, or handling layers/obsm/backed-mode; for the QC, normalization, clustering, UMAP, and differential-expression analysis pipeline prefer alterlab-scanpy instead, and for RNA velocity from spliced/unspliced layers prefer alterlab-scvelo instead. Part of the AlterLab Academic Skills suite.