AI & ML

sboghossian-mini-claude-for-legal Skill

Use when reviewing an existing NDA (mutual or unilateral) to assess whether it adequately protects the disclosing party's confidential information. Evaluates the definition of confidential information, duration of obligations, permitted disclosures, remedies for breach, return/destruction provisions, and residuals clauses. Identifies weaknesses and proposes stronger alternative drafting. Applicable across MENA, EU, UK, and US jurisdictions.

sboghossian-mini-claude-for-legal Skill

Use when drafting a post-hearing brief (also called a post-hearing submission or closing written submission) in an international arbitration or formal hearing. Synthesizes the evidentiary record, addresses witness credibility, analyzes applicable law, responds to opposing arguments, and requests specific relief with supporting reasoning. Applicable to ICC, DIAC, LCIA, SIAC, ADCCAC, and DIFC-LCIA arbitrations as well as common-law court proceedings.

sboghossian-mini-claude-for-legal Skill

Use when an employer needs to draft a remote work policy covering eligibility, equipment, working hours, communication expectations, data security, expense reimbursement, workspace safety, and performance monitoring. Particularly important in MENA jurisdictions where remote work intersects with UAE, Saudi, and Lebanese labour law requirements, visa/residency obligations, and data protection rules.

gptomics Bundle

Judges whether a macromolecular model (or a region of it) is reliable enough to build on, using resolution, R-free, B-factors, MolProbity geometry, and predicted-model confidence with Bio.PDB. Use when deciding if a structure or a specific region is trustworthy before docking/mechanism/measurement; reading resolution, R-work vs R-free and the R-free-minus-R-work overfitting gap; sanity-checking per-residue and mean B-factors; flagging clashscore, Ramachandran and rotamer outliers and cis non-proline peptides; validating a PREDICTED (AlphaFold/ESMFold) model via pLDDT bands and PAE before docking or molecular replacement; and interpreting cryo-EM global-vs-local resolution (FSC 0.143 half-map vs 0.5 map-model) or an NMR ensemble spread. Keywords validation, resolution, R-free, B-factor, MolProbity, clashscore, Ramachandran, rotamer, pLDDT, PAE, wwPDB, cryo-EM local resolution.

gptomics Bundle

Quantifies biodiversity from species abundance/incidence tables using Hill numbers (iNEXT) with coverage-based rarefaction-extrapolation (Chao & Jost 2012), asymptotic richness via Chao1/ACE/jackknife as a lower bound, Baselga turnover/nestedness partition with the Podani alternative as sensitivity check, mandatory Hellinger transformation before ordination (Legendre & Gallagher 2001), Faith PD and SES_MPD/SES_MNTD with explicit null-model choice, and Maire 2015 functional-diversity dimensionality optimization. Use when comparing diversity across sites with unequal sampling effort, picking the right richness estimator for singleton-heavy amplicon data, partitioning beta diversity into turnover vs nestedness, reporting Hill-number effective species counts rather than raw entropies, computing SES_MPD with explicit null-model justification, or deciding whether to apply standard metrics to compositional amplicon data. Not for clinical 16S microbiome diversity (see microbiome/diversity-analysis).

gptomics Bundle

Infer transcription factor regulons from single-cell RNA-seq with pySCENIC by combining GRNBoost2 co-expression, cisTarget motif-enrichment pruning, and AUCell per-cell activity scoring. Covers the motif-pruning-as-directionality principle, regulon specificity scoring, run-to-run stability, and database/species matching. Use when identifying TF regulons, scoring TF activity per cell, finding master regulators of cell identity, or comparing regulon activity across conditions. For enhancer-driven multiomic GRNs see multiomics-grn; for bulk inference and VIPER protein-activity see grn-inference.

gptomics Bundle

Segment single cells from multiplexed IMC/MIBI tissue images using Mesmer/DeepCell, Cellpose, or ilastik+CellProfiler, covering whole-cell vs nuclear segmentation, the summed-membrane-channel decision, nuclear-expansion bias, lateral spillover, resolution-floor parameters, and downstream-proxy evaluation. Use when delineating cells after preprocessing, choosing a segmentation model, building a cell mask for quantification, diagnosing impossible double-positive populations, or troubleshooting over/under-segmentation.

gptomics Bundle

Rank and prioritize neoantigen/epitope candidates by likely T-cell response using NeoFox feature annotation, PRIME2.0, BigMHC-IM, the Łuksza/Balachandran fitness model (agretopicity + foreignness), and pVACtools tiering. Encodes the field's hard truths that immunogenicity is the least-solved layer (dedicated scores ~AUROC 0.6-0.7, modest PPV), that scores are valid only for RANKING within one patient (never absolute go/no-go or cross-patient), that DAI has anchor-inflation and WT-denominator traps, and that stacking weak correlated scores into one number is a red flag. Use when ordering a candidate list for a vaccine. Binding lives in mhc-binding-prediction; calling in neoantigen-prediction.

gptomics Bundle

Aligns Oxford Nanopore and PacBio long reads (and assemblies) to a reference with minimap2 using the error-rate-matched preset (map-ont, lr:hq, map-hifi, map-pb, splice/splice:hq, asm5/10/20, ava), producing a sorted/indexed BAM for variant, SV, methylation, or isoform analysis. Covers why the preset rewrites the scoring/chaining model, why SV calling rides on supplementary not secondary alignments, carrying MM/ML methylation tags through with -y, the multi-part-index MAPQ trap, and when to swap in Winnowmap/VACmap/lra/pbmm2. Use when mapping ONT or PacBio reads, choosing a minimap2 preset by platform/chemistry, preparing input for Clair3/medaka/Sniffles/modkit, aligning into repeats/centromeres, or spliced-aligning cDNA/Iso-Seq.

gptomics Bundle

Discovers shared and view-specific latent factors across bulk multi-omics blocks (RNA-seq, proteomics, methylation) on a common sample axis with MOFA2's unsupervised Bayesian group factor model, then attributes per-view variance explained and interprets signed factor weights. Covers why a factor is an unsupervised axis of variance and not a pathway, why a factor that correlates with batch is a batch factor, why the per-view variance-explained table is the primary read-out rather than p-values, why raw counts in a Gaussian view make factor 1 the library-size factor, and why MOFA2 handles missing omics-per-sample natively. Use when integrating two or more bulk omics to find joint axes of variation, choosing factor count, labeling factors against metadata, or running enrichment on factor weights. For supervised discriminant integration see mixomics-analysis; for the method decision see integration-design; for single-cell see single-cell/multimodal-integration; for enrichment see pathway-analysis/gsea.

gptomics Bundle

Scans genomes for natural selection with SFS tests (Tajima's D, Fay & Wu H, Zeng E, SweepFinder2 CLR), haplotype tests (iHS, nSL, XP-EHH, Rsb, H12), and differentiation (FST, PBS) using scikit-allel, selscan, and SweepFinder2. No single statistic separates selection from demography at one locus, so the deliverable is empirical genome-wide outliers plus multiple orthogonal signals, not an absolute cutoff. iHS detects incomplete sweeps and collapses to zero at fixation while XP-EHH catches fixed sweeps; iHS/nSL standardize within derived-allele-frequency bins but XP-EHH gets a genome-wide z-score; derived-allele tests need substitution-model polarization; background selection mimics FST and CLR. Use when computing selection statistics like FST, Tajima's D, iHS, or XP-EHH, or scanning for selective sweeps. For phasing inputs see phasing-imputation/haplotype-phasing; for dN/dS see comparative-genomics/positive-selection.

gptomics Bundle

Retrieves and interprets AlphaFold Protein Structure Database (AFDB) models by UniProt accession, reading pLDDT and PAE confidence correctly. Use when treating pLDDT as PER-RESIDUE confidence (not global accuracy) and recognizing a long low-pLDDT stretch as an intrinsically disordered region rather than a modeling error; reading PAE to segment confident domains and judge inter-domain/relative-position confidence that high mean pLDDT cannot certify; recognizing a static AFDB model carries NO ligands, ions, cofactors, PTMs, quaternary assembly, or alternative conformations (pLDDT sits in the B-factor column with opposite polarity to thermal motion); and deciding an AFDB entry vs re-running prediction. Keywords AlphaFold DB, pLDDT, PAE, B-factor column, intrinsic disorder, UniProt, Foldseek.

gptomics Bundle

Prepares a deposited or predicted structure for docking, molecular dynamics, or electrostatics by adding hydrogens, assigning protonation and tautomer states, and filling missing atoms and short loops with PDBFixer, reduce, PROPKA, and PDB2PQR. Use when adding hydrogens an X-ray model never resolved; assigning His HID/HIE/HIP tautomers, Asn/Gln/His 180-degree flips, and Cys/Lys/Asp/Glu pKa-shifted protonation at a stated pH and microenvironment rather than trusting standard pKa 7; filling missing side-chain atoms and modeling short missing loops as disorder hypotheses; making a receptor docking- or MD-ready and recording what was built; preparing a predicted model after trimming low-pLDDT regions; and writing a PQR for Poisson-Boltzmann electrostatics. Keywords PDBFixer, reduce, PROPKA, PDB2PQR, protonation, tautomer, missing atoms, hydrogens, pKa, docking prep, MD prep.

sboghossian-mini-claude-for-legal Skill

Use when counsel needs to draft a statement of claim (or memorial, or claim brief) for a client bringing a civil or commercial dispute before a court or arbitral tribunal. Covers factual background, legal causes of action, heads of damage, and relief sought. MENA-specific guidance addresses UAE civil court pleading practice, DIFC/ADGM claim form requirements, Lebanese and Egyptian court procedures, and the distinction between civil-law declaratory and Arabic-language court filings.

sboghossian-mini-claude-for-legal Skill

Use when designing, auditing, or debugging the PostgreSQL Row Level Security (RLS) policies that enforce tenant isolation in the multi-tenant legal AI product. Covers the membership-based isolation model, policy patterns for all core tables, common bypass traps, and the mandatory integration test suite that verifies no cross-tenant data leakage.

sboghossian-mini-claude-for-legal Skill

Use when running or interpreting the competitor comparison benchmark that measures the legal AI system's output quality against comparable tools on standardized legal tasks. Provides structured side-by-side scoring methodology, output dimensions, and the MENA-jurisdiction coverage gap analysis that differentiates this system.

autohandai-community-skills Bundle

Connects NemoClaw to a local inference server. Use when setting up Ollama, vLLM, TensorRT-LLM, NIM, or any OpenAI-compatible local model server with NemoClaw. Trigger keywords - nemoclaw local inference, ollama nemoclaw, vllm nemoclaw, local model server, openai compatible endpoint, switch nemoclaw inference model, change inference runtime, nemoclaw additional model, nemoclaw sub-agent model, openclaw sub-agent, agents.list, sessions_spawn, vlm-demo, nemoclaw tool calling, ollama tool calls, vllm tool-call-parser, raw json in tui, nemoclaw inference options, nemoclaw onboarding providers, nemoclaw inference routing.

autohandai-community-skills Bundle

Identify and eliminate host-device synchronizations in PyTorch code. Detects sync points (.item(), .cpu(), boolean indexing, torch.tensor on CUDA), classifies false vs true dependencies, provides sync-free alternatives. Triggers: sync-free, synchronization, .item(), .cpu(), host-device sync, eliminate syncs, CPU stall, non_blocking, set_sync_debug_mode, cudaStreamSynchronize, cudaEventSynchronize, remove syncs, async GPU.

sboghossian-mini-claude-for-legal Skill

Use when drafting a franchise agreement granting a franchisee the right to operate a branded franchise at a specified location. Covers franchise fees, royalties, training requirements, operational standards, territory restrictions, term, renewal rights, and termination provisions. Applicable across MENA (UAE, KSA, LB, EG) and internationally. Critical in UAE and KSA where franchise arrangements can trigger mandatory commercial agency protection laws if not structured carefully. Trigger when a franchisor is entering a new market or when a franchisee needs its rights documented.

sboghossian-mini-claude-for-legal Skill

Use when migrating the Patrick Munro legal simulation methodology into the mini-claude-for-legal format. This import adapter preserves adversarial scenario modelling — moot-court style argument construction, counterparty position mapping, and litigation outcome simulation — mapping it into the standard skill model. Suitable for dispute-risk assessment, negotiation preparation, and legal education contexts across common-law and civil-law jurisdictions.

sboghossian-mini-claude-for-legal Skill

Use when drafting an AI governance policy for an organization addressing responsible AI use, risk assessment, bias detection, transparency, human oversight, data handling for AI training, vendor evaluation, and incident response. Covers privacy and data protection practice area across MENA (UAE, KSA, EG) and global jurisdictions, with attention to emerging AI-specific regulatory frameworks.

sboghossian-mini-claude-for-legal Skill

Use when a litigator or disputes lawyer needs to produce a structured case assessment memo for a client, covering facts, legal issues, strengths/weaknesses, estimated outcomes and damages, and strategic recommendations. Applicable across MENA civil-law courts (Lebanon, UAE onshore, KSA, Egypt), DIFC/ADGM common-law courts, ICC/DIAC/LCIA arbitration, and any other forum once the jurisdiction is specified.

sboghossian-mini-claude-for-legal Skill

Use when a lawyer needs to draft a professional advisory note explaining the tax implications, compliance requirements, and legal risks of a proposed transaction to a business client. Covers VAT/tax analysis, regulatory compliance, and risk disclosure across MENA jurisdictions (UAE, KSA, LB, EG), DIFC/ADGM, and other applicable legal systems. Calibrated for delivery to sophisticated non-lawyer recipients — clear, structured, no jargon.

sboghossian-mini-claude-for-legal Skill

Use when a company needs to draft a consulting agreement engaging an individual or corporate consultant to provide services. Covers compensation structure, expense reimbursement, IP ownership, confidentiality, non-compete and non-solicitation, and independent contractor status. Addresses MENA jurisdiction traps around employee-vs-contractor misclassification under UAE, KSA, LB, and EG labor law, and distinguishes DIFC/ADGM common-law treatment from civil-law onshore frameworks.

sboghossian-mini-claude-for-legal Skill

Use when a lawyer needs to review a contract and produce a structured risk matrix categorising each clause by risk level (low/medium/high/critical), with clause reference, risk description, potential impact, likelihood, and recommended mitigation. The output is a decision-support tool for clients and deal teams. Applicable to all contract types and jurisdictions; MENA-aware for UAE, KSA, LB, EG, DIFC/ADGM enforcement realities.

sboghossian-mini-claude-for-legal Skill

Use when preparing or summarizing a legal due diligence report on a target company in connection with an M&A transaction. Covers corporate structure, material contracts, regulatory compliance, litigation exposure, IP, real estate, employment, and IT/data. Identifies material risks, assigns risk ratings, and recommends deal protections. Relevant for transactions across MENA (UAE, KSA, DIFC, ADGM, LB, EG), UK, EU, and GCC. Trigger when buy-side counsel needs to consolidate findings from document review into a formal report for the client.

sboghossian-mini-claude-for-legal Skill

Use when a legal department or law firm is issuing a request for proposal (RFP) for a legal technology solution such as a contract management system, e-billing platform, matter management system, or legal research tool. Produces a structured RFP covering functional and technical requirements, security standards, integration needs, evaluation criteria, and pricing. Applicable to corporate legal departments and law firms of all sizes.

sboghossian-mini-claude-for-legal Skill

Use when parties to a dispute need to draft a settlement agreement resolving pending or threatened litigation or arbitration. Covers settlement amount, payment terms, mutual releases, confidentiality, non-disparagement, and dismissal of claims. MENA-specific guidance on release enforceability under UAE civil law, waqf from liability under Lebanese law, confidentiality limitations in KSA court proceedings, and Arabic-language execution requirements.

sboghossian-mini-claude-for-legal Skill

Use when a defendant or respondent needs to draft a statement of defense to a civil or commercial claim, addressing each allegation, asserting applicable defenses, and including any counterclaims. Covers factual admissions and denials, affirmative defenses, counterclaims, and the legal framework for each. MENA-specific guidance covers UAE civil court defense practice, DIFC/ADGM defense procedures, and the strategic differences in civil-law vs. common-law pleading.

sboghossian-mini-claude-for-legal Skill

Use when a company needs to draft or update an internal whistleblower (speak-up) policy covering reporting channels for suspected wrongdoing, types of reportable concerns, investigation procedures, anti-retaliation protections, and confidentiality guarantees. Relevant for corporate governance compliance across MENA and international jurisdictions, especially for publicly listed companies, financial institutions, and multinationals with EU or US operations where mandatory whistleblower frameworks apply.

autohandai-community-skills Bundle

Implement Microsoft's Enhanced Security Admin Environment (ESAE) tiered administration model for Active Directory. Covers Tier 0/1/2 separation, privileged access workstations (PAWs), administrative f

jmagly-ai-writing-guide Skill

Production readiness review — strip prototype scaffolding, harden code, validate cost model, generate prod/ artifacts

jmagly-ai-writing-guide Skill

Manage the AIWG MCP server — start it, configure clients, and maintain the registered server registry

sboghossian-mini-claude-for-legal Skill

Use when migrating a legal risk-assessment skill originally authored for the Anthropic Claude API into the mini-claude-for-legal format. The adapter maps legacy risk-scoring matrices, issue taxonomies, severity buckets, and remediation guidance into the standard skill model. Supports multi-jurisdictional risk frameworks covering MENA (UAE, KSA, LB, EG), EU, UK, and France.

sboghossian-mini-claude-for-legal Skill

Use when migrating the Patrick Munro red-team verification methodology into the mini-claude-for-legal format. This adapter maps adversarial stress-testing logic for legal documents and AI-generated legal output — deliberate challenge of assumptions, identification of counter-arguments, and verification of factual and legal accuracy — into the standard skill model. Triggers when importing any red-team or adversarial-review skill attributed to the Munro approach.

sboghossian-mini-claude-for-legal Skill

Use when drafting a bonus or commission plan for sales, executive, or employee roles covering eligibility, performance metrics, calculation methodology, payment timing, pro-ration, clawback, and discretionary elements. Employment practice area with MENA-specific attention to WPS (Wage Protection System) requirements, gratuity interaction, and Sharia-compliant commission structures in KSA and UAE.