Coding & Dev Tools

jmagly-ai-writing-guide Skill

Configure, use, troubleshoot, or extend AIWG composable output modes (also called output masks), including Wittgenstein-inspired clarity, engineering controlled language, and personal syntax profiles

jmagly-ai-writing-guide Skill

Run a diversity-first premortem with bounded deep dives, blind plausibility and impact verification, and explicit provenance labels.

jmagly-ai-writing-guide Skill

Suggest and apply fixes for flaky tests based on detected patterns. Use after flaky-detect identifies unreliable tests that need repair.

jmagly-ai-writing-guide Skill

Detect orphaned tests, obsolete assertions, and test-code misalignment. Use for test suite maintenance, cleanup, and traceability validation.

jmagly-ai-writing-guide Skill

Extract, classify, deduplicate, and enrich IOCs from investigation artifacts; map to STIX 2.1 observables

sboghossian-mini-claude-for-legal Skill

Use when an attorney or client needs to compare the expected value of accepting a settlement offer against the expected value of proceeding to trial — incorporating P(win), damages range, costs to trial, loser-pays rules, discount rate, and strategic or reputational factors. Produces an EV comparison (trial vs. settlement), a recommendation (take/counter/decline), and a sensitivity analysis showing which assumptions matter most. Always disclaimed as a probabilistic estimate. Always paired with the outcome probability estimator.

sboghossian-mini-claude-for-legal Skill

Use when a user wants to draft a litigation brief, complaint, statement of claim, statement of defense, or any first substantive court pleading, and Claude must gather the jurisdictional, procedural, and factual inputs before generating the document. Triggers on requests to draft, prepare, or review court filings, complaints, or defenses across MENA civil, commercial, Sharia, and free-zone courts, as well as UK, French, and US federal/state jurisdictions.

sboghossian-mini-claude-for-legal Skill

Use when a legal AI conversation exceeds approximately 50 turns or approaches 80% of the context window, and Claude must compress older turns into a structured matter-summary without losing any material decisions, open issues, or user preferences. This is a core behavioral skill that applies to all legal matter types across all jurisdictions. Triggers automatically on context pressure, at major matter milestones, or on explicit user request.

sboghossian-mini-claude-for-legal Skill

Use as a companion to draft-distribution-agreement when the distribution arrangement involves MENA jurisdictions. Provides jurisdiction-specific variant clauses for KSA, UAE, Lebanon, and GCC contexts — including commercial agency registration risk mitigation, UAE Commercial Agencies Law (Federal Law 18/1981) analysis, KSA Saudi-national ownership requirements, Lebanese commercial agency law nuances, and GCC trademark exhaustion considerations.

sboghossian-mini-claude-for-legal Skill

Use when a law firm administrator or practice group leader needs to monitor timekeeper utilization — the ratio of billable hours logged to target hours — across attorneys, practice groups, and time periods. Produces a real-time dashboard with utilization rates, benchmark comparisons, trend lines, and actionable staffing recommendations including burnout risk and under-utilization flags. Benchmarks span US BigLaw, UK City, and MENA practice norms.

sboghossian-mini-claude-for-legal Skill

Use when writing or reviewing copy for the main public homepage of a legal AI assistant. Defines the specific messaging constraints, section-level rules, and claim hierarchy for the homepage surface — which must serve both consumer and professional audiences simultaneously and is the primary surface for both first impressions and regulatory scrutiny. Applied as part of the messaging-compliance-checker pre-publication gate.

techwavedev Skill

AI-powered image editing with style transfer and object removal

techwavedev Skill

Microsoft 365 Agents SDK for Python. Build multichannel agents for Teams/M365/Copilot Studio with aiohttp hosting, AgentApplication routing, streaming responses, and MSAL-based auth.

techwavedev Skill

Microsoft 365 Agents SDK for TypeScript/Node.js.

techwavedev Skill

Design short-term, long-term, and graph-based memory architectures

techwavedev Skill

Create Mermaid diagrams for flowcharts, sequences, ERDs, and architectures. Masters syntax for all diagram types and styling.

techwavedev Bundle

Manage track lifecycle: archive, restore, delete, rename, and cleanup

techwavedev Skill

Expert .NET backend architect specializing in C#, ASP.NET Core, Entity Framework, Dapper, and enterprise application patterns.

techwavedev Bundle

Create distributable Python packages with proper project structure, setup.py/pyproject.toml, and publishing to PyPI. Use when packaging Python libraries, creating CLI tools, or distributing Python ...

autohandai-community-skills Bundle

Validate and use CPU offloading in Megatron Bridge, including layer-level activation offloading and fractional optimizer state offloading with HybridDeviceOptimizer.

autohandai-community-skills Bundle

Cut a new semver release — bump all version strings via bump-version.ts, open a release PR, and after merge tag main and push. Use when cutting a release, tagging a version, shipping a build, or preparing a deployment. Trigger keywords - cut tag, release tag, new tag, cut release, tag version, ship it.

autohandai-community-skills Bundle

Converts cuTile GPU kernels (@ct.kernel) to Triton (@triton.jit). Handles standard in-repo conversion, debugging (cudaErrorIllegalAddress, shape mismatch, numerical mismatch), and mapping cuTile idioms (ct.load/ct.store, ct.Constant, ct.launch) to Triton equivalents. Covers dual-kernel layout flags (e.g. transpose=True/False + autotune grid via META) per translations/advanced-patterns.md. Use when converting, porting, or translating cuTile kernels to Triton, or debugging existing Triton translations.

autohandai-community-skills Bundle

Configure and execute authenticated vulnerability scans using OpenVAS/Greenbone Vulnerability Management with SSH and SMB credentials for comprehensive host-level assessment.

autohandai-community-skills Bundle

Dark web monitoring involves systematically scanning Tor hidden services, underground forums, paste sites, and dark web marketplaces to identify threats targeting an organization, including leaked cre

jmagly-ai-writing-guide Skill

Orchestrate Delivery Track flow with test-driven development, quality gates, and iteration assessment

jmagly-ai-writing-guide Skill

Plan iterations, delegate to SDLC agents, and compile iteration status

jmagly-ai-writing-guide Skill

Validate provenance records and chains for completeness and consistency

jmagly-ai-writing-guide Skill

Create and maintain regression test baselines for comparison and drift detection across versioned snapshots

jmagly-ai-writing-guide Skill

Restore the .aiwg/ directory from a migrate-workspace backup, listing available backups when none is specified

jmagly-ai-writing-guide Skill

Create a new extension package inside an existing framework's extensions/ directory. Use when authoring an expansion (extension) for AIWG.

jmagly-ai-writing-guide Skill

Create a complete framework package structure inside agentic/code/frameworks/. Use when authoring an expansion (framework) for AIWG.

sboghossian-mini-claude-for-legal Skill

Use when deciding how to handle requests to reproduce legal source material — case reports, headnotes, treatise excerpts, statute text, or legal news. Defines which categories of legal content are public domain vs. copyright-protected, the rule against verbatim headnote reproduction (Westlaw/Lexis headnotes are publisher-copyrighted), how to handle statute text from official vs. annotated sources, and the hard refusal for paywall-circumvention requests. Applies across all jurisdictions.

sboghossian-mini-claude-for-legal Skill

Use when a user wants to draft a power of attorney (POA) and Claude must collect the identity, scope, duration, and notarization inputs before generating the instrument. Triggers on requests to prepare a general POA, special POA, limited authority letter, or agency arrangement. Covers MENA notarization requirements (KSA MOJ Najiz, UAE Notary, Lebanon Notary) and cross-border apostille/legalization chains. Routes to draft-power-of-attorney with a notarization checklist.

sboghossian-mini-claude-for-legal Skill

Use when drafting or reviewing governing law and dispute resolution clauses. Enforces the rule that the governing law and the dispute resolution forum must be compatible — or, if intentionally mismatched, that the mismatch is acknowledged and its consequences managed. Mismatched governing law and forum creates enforcement risk, cost, and uncertainty. Applies to all commercial contracts with MENA parties or MENA-seated arbitration. Flags the common anti-patterns including "international law" as governing law and unmatched arbitration/court clauses.

sboghossian-mini-claude-for-legal Skill

Use as a mandatory post-draft quality pass on any legal document or advice output. Requires a dedicated review of every number and date in the output — statute article numbers, case citations, deadlines, limitation periods, currency amounts, percentages, and Hijri/Gregorian dates. AI legal drafting errors are most visible and most damaging when they involve incorrect numbers; this heuristic prevents those failures. Applies universally across all jurisdictions and document types.

sboghossian-mini-claude-for-legal Skill

Use whenever a user is preparing documents for court filing, government registration, cross-border authentication, or any official purpose in a MENA jurisdiction that requires translation from or into Arabic or another language. Defines the four levels of translation authority (machine, self-certified, certified, sworn), maps jurisdiction-specific requirements for LB/KSA/UAE/EU, identifies the end-use determination step, and flags the common error of commissioning the wrong translation authority level.