major-ai-skills

Plugin major-ai-skills from marketplace major-ai-skills in alivirgo/Major-AI-Skills. Commands read from the repo's manifest — yesterday.

Skills (200)

• 3d-slicer

  Automate 3D Slicer research workflows with MRML, VTK/ITK, DICOM data, and Segment Editor. Use for software workflows, not clinical interpretation.

• abbreviated-git-status

  Inspect Git branch, staging, and working-tree changes with short or porcelain status output when a compact repository check is needed.

• ableton-live

  Automate Ableton Live sessions with Python MIDI Remote Scripts, Max for Live, AbletonOSC, and Link synchronization.

• act-as-a-personal-tutor

  Explain a subject step by step, assess prior knowledge, and use short exercises to check understanding before advancing.

• add-your-own-human-voice

  Revise AI-assisted writing to preserve the author's voice, concrete experience, and intended tone.

• adobe-after-effects

  Scaffold After Effects comps and expression rigs, queue Render Queue / AME jobs, and export Essential Graphics (.mogrt) with ExtendScript-safe patterns.

• adobe-photoshop

  Automate Photoshop with UXP or ExtendScript, Actions and Batch/Image Processor, Smart Object pipelines, and color-managed exports—catching profile, bit-depth, and headless-DOM failures agents routinely miss.

• adobe-premiere-pro

  Automate Premiere Pro sequences and markers via UXP/ExtendScript, exchange FCPXML/AAF/EDL, and queue Adobe Media Encoder jobs without baking secrets into projects.

• agent-injection-boundary-test

  Test authorized agents against indirect prompt injection in retrieved docs and tool outputs using paired clean/poison fixtures, synthetic canaries, and sandboxed tools. Evidence for these cases—not a claim injection is solved.

• agent-tool-replay-test

  Replay recorded agent tool/LLM trajectories against deterministic fixtures to test argument validation, errors, idempotency, and side-effect boundaries—separate from live tool-selection quality.

• agentic-token-efficiency-master-rule

  Plan context use, bounded file inspection, tool output, and response length for coding-agent tasks with an explicit token budget.

• ai-citation-verification

  Verify AI citations across structural schema, resolvability (registry/URL/DOI), and semantic support of each claim by the cited span—not the whole context. Use for RAG answers, research agents, and regulated audit trails.

• ai-evaluation-dataset

  Build versioned JSONL evaluation datasets for AI workflows with golden/regression/holdout layers, leakage checks, rubrics, and provenance hashes. Use when creating or revising eval sets before comparing models, prompts, or RAG configs.

• ai-human-handoff-contract

  Define and test AI→human handoff as a versioned state machine with multi-signal triggers, review packets, approval binding, and no silent consent. Use for escalations, approvals, and irreversible agent actions.

• ai-pii-redaction-review

  Privacy-review AI payloads with field-aware redaction, typed stable placeholders, leakage scans across prompts/logs/tools/RAG, and residual-risk reporting before external submission. Use before sending data to cloud LLMs or sharing traces.

• airflow

  Build and troubleshoot Apache Airflow DAGs, operators, sensors, schedules, and retry behavior for data pipelines.

• altium-designer

  Automate Altium Designer PCB workflows with DXP scripts, design-rule checks, OutJob fabrication outputs, and routing configuration.

• always-verify-numbers-and-links

  Verify statistics, contact details, calculations, and URLs against appropriate sources before using them in a deliverable.

• android-studio

  Configure Android Studio Gradle projects, emulators, build variants, and signing; diagnose build and runtime failures with AGP and Logcat.

• ansi-strip-logger

  Remove ANSI colors and terminal control sequences from logs before passing them into model context or structured reports.

• ansible

  Write idempotent Ansible playbooks, FQCN modules, roles, and inventories; gate prod with check/diff, pin collections, and keep ansible-vault secrets out of Git and chat.

• anthropic-api

  Integrate Anthropic Messages API with pinned Claude models, required max_tokens, tool_use loops, streaming, and prompt caching agents routinely miswire.

• appcleaner

  Inspect macOS application bundles and related files for AppCleaner workflows, review removal candidates, and troubleshoot SmartDelete behavior.

• arcgis-pro

  Automate ArcGIS Pro geoprocessing with ArcPy, enterprise geodatabases, and Spatial Analyst workflows.

• argocd

  Operate Argo CD GitOps with AppProjects, safe automated sync/prune, and RespectIgnoreDifferences—debug OutOfSync/Degraded without kubectl drift that fights the controller.

• artifact-reuse-pattern

  Update an existing plan or state artifact in place when the same document needs to track progress across multiple agent turns.

• asana

  Organize Asana projects and tasks, configure custom fields and rules, and automate scoped updates through its API.

• ask-ai-confidence-rating

  Identify uncertainty, assumptions, and claims needing verification when reviewing an AI answer.

• ask-ai-to-audit-itself

  Review a draft answer for logical errors, missing requirements, edge cases, and claims needing external verification.

• ask-ai-to-check-its-math

  Check arithmetic with explicit formulas and a calculator or executable code before presenting numerical results.

• ask-ai-to-repeat-understanding

  Restate the intended outcome, scope, and constraints when a complex task needs an alignment check.

• ask-ai-to-show-its-reasoning

  Request a concise explanation of assumptions, evidence, decision criteria, and trade-offs behind a recommendation.

• ask-for-alternative-viewpoints

  Compare opposing viewpoints and stress-test the assumptions behind an argument or recommendation.

• ask-for-tables-over-paragraphs

  Request a table when options, attributes, or numerical results are easier to compare in aligned columns.

• ask-for-three-options

  Compare conservative, balanced, and ambitious approaches with their trade-offs when choosing a course of action.

• ask-how-to-improve-prompt

  Review a prompt for missing context and ambiguous requirements, then produce a reusable revised version.

• ask-one-best-answer

  Request one recommended option with its rationale when a long list would obscure the decision.

• ask-short-answer-first

  Lead with a brief direct answer and place supporting detail afterward when the reader needs a quick decision.

• ask-what-changed-only

  Request only changed lines or sentences and their relevant context when reviewing an update.

• ask-what-info-is-missing

  Identify missing inputs and ask targeted questions before resolving an underspecified task.

• ast-refactoring-scripts

  Use AST-aware tools such as libcst, ts-morph, or jscodeshift for repeatable code migrations across many files.

• ast-skeletonizer

  Extract types, classes, and function signatures into a compact code outline when investigating structure before implementation details.

• atomic-code-edits

  Apply localized code replacements when only a small part of a file changes, preserving surrounding implementation and review context.

• autohotkey-v2

  Write AutoHotkey v2 desktop scripts for window management and Win32 integration; troubleshoot permissions and compile scripts with Ahk2Exe.

• autopsy

  Automate authorized forensic analysis in Autopsy with Sleuth Kit ingest modules, evidence-processing pipelines, and search indexing.

• avoid-duplicate-system-messages

  Consolidate repeated system instructions into a stable prompt prefix when configuring an application's prompt caching.

• avoid-rephrasing-good-text

  Edit awkward or incorrect passages while preserving wording, voice, and humor that already work.

• aws-cli

  Operate AWS CLI v2 with IAM Identity Center SSO, OIDC/CI roles, and account-scoped S3/EC2/Lambda/IAM/CloudWatch workflows agents routinely mis-target.

• azure-cli

  Operate Azure CLI (az) with subscription-scoped ARM, managed identities, AKS/App Service/Key Vault workflows, and RBAC patterns agents mis-scope.

• batch-api-job-processing

  Organize asynchronous evaluation, audit, or dataset workloads for a provider's batch API when immediate responses are unnecessary.

• batch-file-creation

  Scaffold related files with a local generator when a repeatable multi-file layout is needed.

• batch-small-questions-together

  Group related short questions into one structured request while keeping each answer identifiable.

• be-specific-instead-of-vague

  Replace vague requests with a concrete goal, relevant context, constraints, and desired output.

• beware-of-ai-recommendation-bias

  Review product recommendations for popularity bias, affiliate influence, missing alternatives, and unsupported comparisons.

• beware-of-hallucinated-quotes

  Verify quotations, book titles, and academic references against original sources before publishing them.

• binary-file-suppression

  Detect binary, compiled, and media files before text ingestion so agent context contains readable source and metadata.

• blender

  Automate Blender scenes with the bpy Python API, build Geometry Nodes workflows, configure Cycles or EEVEE renders, and troubleshoot headless batch jobs.

• boolean-flag-responses

  Design boolean or enum outputs for narrow diagnostic and triage tasks whose full result fits a fixed set of values.

• brainstorm-presentation-angles

  Develop distinct opening angles and narrative approaches for a presentation, pitch, or keynote.

• break-big-questions-into-steps

  Divide a complex question into manageable subtasks with clear dependencies and intermediate results.

• break-complex-projects-into-milestones

  Turn a project goal into milestones with deliverables, dependencies, and verifiable completion criteria.

• browser-subagent-task-bounding

  Define a browser subtask's target, step budget, completion criteria, and return format before delegating navigation.

• bruno

  Create version-controlled Bruno API collections, write Bru request files, and run API checks through the Bruno CLI in CI.

• build-your-ai-common-sense-habit

  Build a repeatable prompting habit around clear goals, relevant context, verification, and human review.

• bullet-point-summary-first

  Extract a short action-focused bullet summary from a long document, article, or meeting record.

• cached-system-instructions

  Structure stable system instructions and variable input separately when configuring supported prompt-cache workflows.

• cached-tool-permission-grants

  Reuse session-scoped tool approvals where the host supports them, preserving the user's approved scope and existing permission boundaries.

• cadence-virtuoso

  Automate Cadence Virtuoso design and simulation with SKILL, OCEAN, and Spectre; inspect DRC and LVS verification results.

• catia

  Automate CATIA V5 and 3DEXPERIENCE modeling workflows through CAA or COM, including shape design and assembly operations.

• check-code-for-obvious-flaws

  Review generated code for nonexistent APIs, exposed secrets, unhandled errors, and likely edge-case failures.

• chromadb

  Build and operate Chroma collections for RAG: PersistentClient vs HttpClient, embedding-function contracts, metadata/where filters, and dimension-safe upsert/query paths agents routinely break.

• cinema-4d

  Script Cinema 4D scenes with c4dpy and the Maxon API, wire MoGraph/Fields systems, and debug Redshift GPU renders, AOVs, and headless Commandline farm jobs agents routinely miss.

• cite-specific-section-numbers

  Anchor document analysis to exact sections or pages so readers can verify the supporting passage.

• clean-messy-notes-to-action-items

  Turn unstructured notes into action items with owners, deadlines, decisions, and unresolved follow-ups.

• clickhouse

  Design ClickHouse MergeTree ORDER BY and partitions; batch inserts; use projections and skip indexes; tune mutations, replication lag, and backups without treating it as OLTP.

• cloudflare-dns

  Manage Cloudflare DNS with correct proxy (orange/grey) mode, TTL cutovers, zone backups, and scoped API tokens—avoid mail/ACME breakage and mixed proxied-record traps.

• codeql

  Configure GitHub CodeQL analysis, run or write queries, interpret SARIF findings, and verify fixes for reported vulnerabilities.

• codesys

  Develop CODESYS Structured Text and Python ScriptEngine workflows; troubleshoot fieldbus and OPC UA integration in a test environment.

• combine-human-speed-with-ai-scale

  Plan work that uses AI for drafting and synthesis while retaining human judgment over priorities and final decisions.

• comment-stripping-preprocessor

  Prepare compact source excerpts for model input by removing redundant comments while retaining necessary context and keeping original files intact.

• compact-dir-listing
• compact-enum-mappings
• compact-error-codes
• compact-error-handling
• compact-guard-clauses
• compact-status-updates
• compact-yaml-frontmatter
• compare-two-products-fairly
• compressed-json-repair
• compressed-system-prompts
• comsol
• concise-output-enforcer
• confluence
• consul
• context-pruning-directive
• convert-voice-memos-to-bullets
• copy-relevant-paragraph-only
• cross-check-legal-advice
• cryengine
• csv-over-json-tables
• cubase
• davinci-resolve
• dbt
• dead-code-elimination
• deduplicated-file-caching
• dense-shorthand-notation
• dependabot-config
• direct-tool-selection
• django
• docker
• dont-repaste-entire-document
• dont-summarize-without-a-goal
• dont-trust-ai-medical-diagnoses
• double-check-ai-math-manually
• draft-emails-with-placeholders
• draft-meeting-agendas
• draft-multiple-variations
• dropover
• duckdb
• dynamics-nav
• early-exit-evaluation
• eartrumpet
• elasticsearch
• encase
• everything
• exit-code-fast-checking
• express
• fastapi
• ffmpeg
• figma
• fill-in-missing-assumptions
• final-cut-pro
• firebase
• fl-studio
• flag-missing-facts-first
• flag-speculative-statements
• flask
• flow-launcher
• flutter
• fluxcd
• format-data-to-copy-paste-tables
• ftk
• fusion-360
• gcloud-cli
• generate-social-media-captions
• github-actions
• github-packages-npm
• gitignore-respecting-search
• give-ai-direct-feedback
• give-ai-situation-context
• give-examples-first
• go-services
• godot
• gpg-signing
• grafana
• graphql-apis
• graphql-client
• graphql-introspection-trimmer
• great-expectations
• gzip-compressed-http-payloads
• head-tail-log-clipping
• header-only-c-cpp-ingestion
• helm
• highlight-action-items
• horos
• houdini
• hubspot
• huggingface-transformers
• ibm-maximo
• in-place-file-mutation
• incident-runbooks
• incremental-diff-ingestion
• inline-lambda-functions
• inline-type-signatures
• iterate-step-by-step
• jira
• json-minified-payloads
• jupyter
• kafka
• keda
• keep-instructions-direct
• keep-instructions-friendly-and-clear
• keep-sensitive-financials-private
• kicad
• kotlin
• kubernetes
• labview
• langchain
• lazy-dependency-loading
• learn-new-skills-ten-times-faster
• lets-encrypt
• line-bounded-file-reading
• linear
• lint-auto-fix-delegation
• list-comprehension-compression
• llamaindex
• llm-cost-latency-benchmark
• llm-json-contract-check
• lm-studio-local-offloading
• logic-pro