AI & ML
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
-
sboghossian-mini-claude-for-legal Skill Tool PDF ExtractorUse as the first step whenever a user uploads a PDF document that needs text analysis, review, or RAG indexing. Extracts text from PDFs with a selectable text layer using PDF.js (fast, no OCR), preserving page boundaries, paragraph structure, headings, and tables for downstream citation. Falls back to OCR tools for scanned documents. Critical to get right — downstream analysis quality depends entirely on clean text extraction.
-
oimiragieo Bundle Security ArchitectSecurity architecture and threat modeling. OWASP Top 10 2025 analysis, OWASP Agentic AI Top 10 (ASI01-ASI10), AI/LLM security patterns, supply chain security, modern API authentication (OAuth 2.1, DPoP, Passkeys/WebAuthn), vulnerability assessment, and security review for code and infrastructure.
0 -
oimiragieo Bundle Swarm CoordinationMulti-agent swarm coordination patterns. Orchestrates parallel agent execution, manages agent communication, handles task distribution, and coordinates results aggregation.
0 -
oimiragieo Bundle Checklist GeneratorGenerate context-aware quality checklists for code review and QA using IEEE 1028 base standards plus LLM contextual additions
0 -
ariffazil Bundle Cloudflare Email ServiceSend and receive transactional emails with Cloudflare Email Service (Email Sending + Email Routing). Use when building email sending (Workers binding or REST API), email routing, Agents SDK email handling, or integrating email into any app — Workers, Node.js, Python, Go, etc. Also use for email deliverability, SPF/DKIM/DMARC, wrangler email setup, MCP email tools, or when a coding agent needs to send emails. Even for simple requests like "add email to my Worker" — this skill has critical config details.
-
ariffazil Skill MCP Zen AuthoringKeep MCP servers minimal, schema-first, and governance-free
-
ariffazil Skill Substrate Propose SealSubstrate primitive /propose-seal — proposes a sealed candidate to 888-APEX. NEVER self-seals. The flow: agent proposes → 888 judges → F13 authorizes → 999 executes (append to VAULT999). Parameterized across Hermes, OpenClaw, and OpenCode runtimes.
-
ariffazil Bundle Archive Vault SealSeal action to civilizational memory with mandatory end-of-session recursive MCP stack hardening and future INIT scaffold. VAULT999 record. Stage 7
-
ariffazil Skill Arif Init WrapperUniversal session auto-anchor for ALL agent vendors on arifOS VPS — Codex, Claude, Gemini, Kimi, OpenCode, Aider, Copilot, and future agents
-
ariffazil Skill MCP Builder DoctrineCanonical MCP server-builder doctrine — naming laws, capability declaration discipline, _meta envelope contract, lifecycle notification rules, error envelope contract, and
-
ariffazil Skill Kernel Verbs Aforge HandsCanonical geometry for arifOS Kernel Verbs vs A-FORGE actuators vs MCP envelope. Eureka 2026-07-08: Kernel Verbs = 12 constitutional primitives (law, zero host mutation). A-FORGE tools = execution actuators (hands, host impact after SESSION+LEASE+GATE+SEAL). MCP "tools" = transport envelope only. Access unrestricted logical; deny via gates not hide. Trigger: session init, "kernel verbs", "hands vs law", forge vs arifOS, MCP tool confusion.
-
gptomics Bundle Bio Phylo Species TreesEstimates species trees under the multispecies coalescent from per-locus gene trees with the modern ASTER astral binary (ASTRAL-III/wASTRAL/ASTRAL-Pro), plus SVDQuartets, BPP, and StarBEAST2. Covers why a species tree is not a gene tree, why each locus has its own genealogy that disagrees by incomplete lineage sorting (ILS) even with zero error, why concatenation is statistically inconsistent and positively misleading in the anomaly zone where more loci converge on the wrong tree with full support, why gene-tree estimation error biases summary methods, that localPP is not bootstrap and ASTRAL branch lengths are coalescent units, and how minority-quartet symmetry separates ILS from introgression. Use when multi-locus discordance, rapid radiations, anomaly-zone risk, concordance-factor interpretation, or concatenation-vs-coalescent choice arise. Routes per-locus gene-tree inference and gCF/sCF to modern-tree-inference, dating to divergence-dating, orthology to comparative-genomics/ortholog-inference.
-
gptomics Bundle Bio Single Cell Cell AnnotationAutomated reference-based cell type annotation for single-cell RNA-seq using CellTypist, SingleR, Azimuth, scANVI, and scmap to transfer labels from a reference. Use when annotating cell types from a reference atlas or pretrained model, transferring labels onto a query, assessing prediction confidence and rejection, or triaging whether an unexpected cluster is a novel type versus a doublet, low-quality, or batch artifact.
-
gptomics Bundle Bio Alignment TrimmingTrim multiple sequence alignments using ClipKIT, trimAl, BMGE, Divvier, or HMMcleaner with mode selection guidance per downstream goal. Use when removing unreliable columns or contaminating residues before phylogenetic inference, HMM building, or selection analysis.
-
gptomics Bundle Bio Alignment MultiplePerform multiple sequence alignment using MAFFT, MUSCLE5, ClustalOmega, or T-Coffee. Guides tool and algorithm selection based on dataset size, sequence divergence, and downstream application. Use when aligning three or more homologous sequences for phylogenetics, conservation analysis, or evolutionary studies.
-
oimiragieo Bundle Gemini CLI SecurityAI-powered code vulnerability analysis and dependency scanning using Gemini CLI security extension patterns. Detects hardcoded secrets, injection attacks, weak cryptography, authentication flaws, and LLM prompt injection. Also scans dependencies against the OSV.dev vulnerability database.
0 -
ariffazil Bundle Ocr And DocumentsExtract text from PDFs/scans (pymupdf, marker-pdf, document_ingest MCP).
-
ariffazil Bundle Hermes AgentConfigure, extend, or contribute to Hermes Agent.
-
ariffazil Bundle Fabrication PreventionArtifact fabrication prevention — verify before claiming existence. Triggered when agent claims file/database/API/artifact existence without external validation.
-
ariffazil Skill Asi MCP GovernorGoverned MCP routing via arifOS F1-F13. All calls to GEOX, WEALTH, WELL organs must route through arifOS kernel for C2+/IRREVERSIBLE tools. Uses mcporter for stdio/HTTP bridging.
-
ariffazil Bundle Unified Memory FederationSingle entry point for memory across the arifOS federation. Routes recall/write/seal/consolidate/audit to the correct tier and organ. Enforces the 6-tier memory model from zen-organ-memory.
-
oimiragieo Bundle Stale Module PrunerRipgrep-powered dead-code crawler that finds stale, broken, or orphaned JavaScript/CJS/MJS modules in the Agent Studio framework. Walks a target directory, checks each file for external references, and surfaces (or deletes) unreferenced files. Prevents registry bloat and silent tool dropouts from modules no one calls.
0 -
oimiragieo Skill Creator Fix WorkerFixes the skill-creator and agent-creator systems (SKILL.md refactoring, hook fixes, template fixes)
0 -
oimiragieo Skill Routing Fix WorkerFixes bugs in the agent-studio routing system (hooks, lib, state management)
0 -
gptomics Bundle Bio Metabolomics Pathway MappingMaps metabolomics results to biological pathways via over-representation (ORA), metabolite-set enrichment (MSEA/QEA), mummichog/PSEA on raw m/z peaks, and network-diffusion enrichment (FELLA), with correct background-set construction and honest interpretive ceilings. Use when interpreting differential metabolites or an untargeted LC-MS feature table in pathway context, choosing ORA vs MSEA vs mummichog vs topology, or setting the reference/background set. For annotation confidence levels feeding ORA see metabolomics/metabolite-annotation; for gene-set concepts see pathway-analysis/go-enrichment and pathway-analysis/gsea; for joint gene+metabolite pathways see multi-omics-integration/mofa-integration.
-
gptomics Bundle Bio Metagenomics Strain TrackingResolves and compares bacterial strains below the species level from shotgun metagenomes with inStrain (popANI/conANI microdiversity), StrainPhlAn (marker-SNV consensus phylogeny and nGD), MIDAS2, metaSNV, and StrainGE, plus genome-vs-genome ANI (skani/fastANI/MASH) for isolate/MAG comparison. Covers why a strain is a threshold not a thing, why ANI answers same-genome while popANI/nGD answer same-population-in-situ, the 99.999% popANI and per-species nGD definitions, the coverage detection limit (absence is not absence), why sharing is not transmission direction, and mapping to the dataset's own dRep MAGs. Use when detecting shared strains, tracking transmission, resolving within-host strain dynamics, or deconvoluting co-occurring strains. For pure-culture isolate outbreak SNP trees see epidemiological-genomics; for MAG assembly see genome-assembly/metagenome-assembly.
-
gptomics Bundle Bio Proteomics Protein InferenceGroups proteins from peptide identifications and controls protein-level FDR, framing inference as a chosen explanation (parsimony or a probability model) of underdetermined peptide evidence rather than a measurement. Reports protein GROUPS (proteins indistinguishable by observed peptides) with a leading protein, not flat lists. Covers shared-vs-unique peptides, indistinguishable/subsumable proteins, parsimony vs probabilistic (ProteinProphet, EPIFANY) vs razor inference, picked-protein and picked-group FDR, and why the two-peptide rule is wrong. Use when resolving which proteins are present from a peptide list, building protein groups, or estimating protein-level FDR. PSM/peptide FDR and search engines are peptide-identification; razor-vs-unique quant consequences are quantification; isoform/proteoform resolution is top-down and out of scope.
-
sboghossian-mini-claude-for-legal Skill Eng MCP Tool RegistryUse when designing, implementing, or maintaining the MCP (Model Context Protocol) tool registry for a legal AI product. Defines how tools are registered, described, scoped, versioned, and discovered by the skill router; covers the tool schema, permission model, legal-product-specific safety constraints, and failure handling. Engineering skill foundational to all skill-to-tool integrations in the eFirm and other suites.
-
ariffazil Bundle Asi Presence OpenSingle entry point for ANY agent spawning in the arifOS federation. Chains existing constitutional skills into an auto-loading sequence. Fixes the gap: "subagents inherit blindly from parent." Every agent loads this ONCE at wake. If any step fails, agent refuses work.
-
sboghossian-mini-claude-for-legal Skill Connector StripeUse when the legal-AI platform needs to manage billing, subscriptions, invoices, or payment lifecycle events — including plan upgrades from the UI, in-chat upsell flows, invoice generation with VAT compliance for UAE and KSA, failed payment recovery, and trial conversion. P0 connector for platform revenue. Covers all subscription tiers (Free through Enterprise) and the add-on credit model. Triggers on any billing, payment, upgrade, refund, or subscription state change request.
-
sboghossian-mini-claude-for-legal Skill Efirm Conflict CheckUse before accepting any new matter to detect conflicts of interest across six dimensions — adverse parties, current clients, former clients (substantially related test), issue/positional conflicts, personal interests, and material-limitation conflicts. Covers ABA Model Rules 1.7–1.10 (US), LB Bar Code of Ethics, KSA Code of Law Practice, UAE Law Society rules, DIFC Rules of Conduct, and ADGM Rules of Conduct. Returns a structured clean/concern/conflict result with waiver guidance. P0 — must complete before substantive work begins.
-
gptomics Bundle Bio Metabolomics Normalization QcDesigns QC, corrects signal drift, removes batch effects, filters features, normalizes samples, and imputes missing values for untargeted LC-MS/GC-MS metabolomics, framing each step as a measurement model that can create or erase biological signal. Use when processing a peak/feature table before statistical analysis, choosing a drift-correction or sample-normalization method, deciding QC RSD vs D-ratio filtering, or handling left-censored missing values. The feature table is produced by metabolomics/xcms-preprocessing or metabolomics/msdial-preprocessing; transformation/scaling for modeling defers to metabolomics/statistical-analysis; cross-study design issues link to experimental-design/batch-design.
-
gptomics Bundle Bio Pathway WikipathwaysTests a gene list (ORA, enrichWP) or a ranked gene vector (GSEA, gseWP) against the WikiPathways community-curated pathway collection with clusterProfiler and rWikiPathways. Covers why a WikiPathways result is a snapshot of a live, monthly-updated database (enrichWP/gseWP/gson_WP silently pull data.wikipathways.org/current/), why reproducibility requires pinning a dated GMT via downloadPathwayArchive(date=, format='gmt'), why the WP GMT is Entrez-keyed so symbols and Ensembl silently overlap nothing, why universe=NULL gives a biased all-WP-genes background, how to split the name%version%wpid%org term, and why WikiPathways (CC0, no peer review) complements KEGG/Reactome. Use when running open community-pathway enrichment, covering a non-model WP species, catching disease/drug pathways missing from KEGG/Reactome, or needing a reproducible dated analysis. The gene list comes from differential-expression/de-results; visualize with enrichment-visualization.
-
gptomics Bundle Bio Single Cell Batch IntegrationIntegrate multiple scRNA-seq samples or batches with Harmony, scVI/scANVI, Seurat (CCA/RPCA), fastMNN, Scanorama, or BBKNN. Resolves which method to use for the dataset size and design, how strongly to correct, when integration is the wrong move (confounded batch/biology), how to score integration with scIB metrics without gaming them, and why corrected expression must not be used for differential expression. Use when integrating batches or datasets, choosing an integration method, diagnosing over-correction, or judging integration quality.
-
oimiragieo Bundle Webmcp Browser ToolsWebMCP — browser-side API that lets web applications expose their own functionality as MCP tools TO AI agents. Use when designing or integrating with web apps that surface UI actions (forms, buttons, data queries) as callable agent tools. NOT for web scraping or fetching external pages.
0 -
oimiragieo Skill Prompt Cache WorkerPrompt assembler optimization — sorting, memoization, cache-break detection, integration tests
0
Frequently asked questions
What are AI & ML agent skills?
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
Which AI & ML skills are most installed?
Popular AI & ML skills on SkillMD right now include mcp-zen-authoring, substrate-propose-seal, ARCHIVE-vault-seal. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do AI & ML skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.