AGENT SKILLS · ÉQUIPEMENT DE MISSION
Des Agent Skills utiles classés selon leur usage réel
Comparez les Skills vérifiés par fonction, contenu, installations et fiabilité de la source avant de les ajouter.
- PÉRIMÈTRE
- VÉRIFIÉ
- Skills
- 22 446
- dépôts
- 3448
- dernière synchronisation
- 29 oct. 2026
bateau84terraform-provider-resources
Implement Terraform Provider resources and data sources using the Plugin Framework. Use when developing CRUD operations, schema design, state management, and acceptance testing for provider resources.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84terraform-provider-testing
Terraform provider acceptance test patterns using terraform-plugin-testing with the Plugin Framework. Covers test structure, TestCase/TestStep fields, ConfigStateChecks with custom statecheck.StateCheck implementations, plan checks, CompareValue for cross-step assertions, config helpers, import testing with ImportStateKind, sweepers, and scenario patterns (basic, update, disappears, validation, regression), and ephemeral resource testing with the echoprovider package. Use when writing, reviewing
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84tui-design
Design, build, refactor, or review terminal interfaces: full-screen TUIs, interactive prompts, and command-line tools. Covers terminal constraints, keyboard interaction, focus, key bindings, color, accessibility, and framework implementation with Bubble Tea, Ratatui, Textual, or Ink. Use for terminal-app layout and UX, CLI behavior, ncurses-style tools, dashboards, REPLs, fzf-like pickers, and requests naming lazygit, k9s, btop, helix, or yazi. Not for browser/web UI, native GUI, editor/font con
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84typescript-async
Asynchronous TypeScript correctness: Promise ownership, awaiting/returning work, async collection patterns, scheduler-sensitive state, cancellation with AbortSignal, cleanup, Promise combinator semantics, timeouts, bounded concurrency, and rejection propagation. Use when a TypeScript change starts, coordinates, cancels, races, retries, or observes asynchronous work.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84typescript-testing
TypeScript/JavaScript testing mechanics across Bun test, Vitest, Jest, and similar runners: awaited async assertions, deterministic timers and concurrency, module/global mock isolation, runtime-vs-type-level tests, environment selection, regression tests, targeted mutation testing, and false-green avoidance. Use when writing or reviewing tests for TypeScript/JavaScript code. Pair with test-driven-development for red-green-refactor process.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84user-flow-design
Task-level state sequences — the path a user takes through steps, decisions, branches, and outcomes across a task or workflow. Use when users must move through more than one view or state to accomplish a goal; when entry points, decision points, or error recovery paths matter; when designing CLI pipelines or TUI navigation between views. Not for per-action response within a single view (→ interaction-design), structural organization of content or commands (→ information-architecture), or cross-c
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84user-stories-and-scenarios
Capture small human-centered User Stories and concrete end-to-end design Scenarios (including a meaningful failure, interruption, or edge condition) from agreed product intent. Use for durable US/SC design artifacts. Not for system requirements or measurable quality scenarios (use behavioral-spec), or implementation design.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84visual-companion
Browser-based visual brainstorming companion for showing mockups, diagrams, and interactive A/B/C choices during conversation. Use when a question would be clearer shown than described — UI mockups, wireframes, architecture diagrams, side-by-side comparisons, flowcharts.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
bateau84web-ui-design
Web-specific interface design: browser behavior, URLs as UX, document vs app models, forms, responsive interactions, and accessibility patterns. Use for SPAs, multi-page apps, forms, dashboards, and admin interfaces. Add visual-interface-design for visual reasoning. Not for CLI design, TUI design, native desktop GUI, or pure content sites with no interaction beyond navigation.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 18 oct.
ilyasemenovtop-down-review
Code review that starts from the task, not the diff. Reviews a branch, MR/PR, or diff in 3 steps: WHY the change matters and whether its product contract is right, WHAT the solution should be, designed from the task, and HOW the code implements it — correctness, tests, documentation, naming, and prose. Use when the user asks to review changes.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 20 oct.
kessleriolocal-whisper
LOCAL voice transcription using OpenAI Whisper. 100% private - audio never leaves your machine. Supports 97+ languages.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 juil.
kusoldsimplified-technical
Write or rewrite text in Simplified Technical English at roughly 80% ASD-STE100 strictness. Use when the user asks to "simplify this", "rewrite for clarity", "explain like an aircraft manual", "use STE / ASD-STE100 / simplified technical English", "de-jargon this", "make this unambiguous", or wants plain-language technical documentation. Not for creative or marketing copy.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 20 oct.
kusoldwiki-lint
Periodic health-check for an LLM Wiki markdown knowledge base. Use when the user says "lint the wiki", "health check the wiki/knowledge base", "clean up the wiki", "check the wiki for contradictions/orphans", "what's stale in my wiki", "wiki maintenance", or similar.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 20 oct.
mroxsovectorcraft
Create and edit vector artwork, paths, text, artboards, and live effects in the VectorCraft application using its control protocol, MCP server, or headless CLI. Use when the user asks to work in VectorCraft.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 14 oct.
pronovixtestrunner
Parallel PHPUnit test runner and binary provisioner. Primary tool for all directory, suite, and batch test execution, including dynamic binary download and PATH resolution when missing.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 16 oct.
sshvarzman-salesforceassigning-sra-via-flow
Wire the 'Define Your Multi-Agent Assignment Criteria' autolaunched flow that dynamically assigns which ServicePlanner Service Rep Assistant (SRA) is shown on a live VoiceCall / MessagingSession / Case, based on that record's field values. This is the OPTIONAL flow you attach in the SRA setup step 'Define Your Multi-Agent Assignment Criteria' (only needed when an org has MORE THAN ONE SRA). The whole contract is a tiny AutoLaunchedFlow: one INPUT text variable named exactly recordId (the interac
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforceauthoring-record-summary-prompt-templates
Build a RECORD SUMMARY type Salesforce prompt template (GenAiPromptTemplate type=einsteingptrecordSummary) that renders an AI briefing on a record's page / Record Summary component / agent record-summary — grounded in that record AND related records (parent record, parent VoiceCall/MessagingSession transcript, Contact, Account) via a Capability PromptFlow data provider. Covers how a record-summary template DIFFERS from a flex template: type=einsteingptrecordSummary, a Object binding, an SObject
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforcebuilding-nba-conversation-intelligence
Use this skill when the user needs to build a Next Best Action (NBA) system triggered by Conversation Intelligence (CI) Signal Rules on a messaging or voice channel. Covers the full stack: screen flows (recommendation actions), Recommendation records, Conversation Intelligence Signal Rules (with keywords), and the Recommendation Strategy Flow that connects rules to recommendations. Trigger when the user mentions NBA, Next Best Action, recommendation, conversation intelligence, signal rules, keyw
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforcebuilding-system-context-agent-data-flows
Build the backing data flows for an Agentforce Service Agent (ASA) so they read and write records in system context regardless of the agent user's sharing/perms. Covers the read pattern (prompt-template-over-PromptFlow + system-context subflow), the write pattern (autolaunched flow action), and the critical empty-results bug where a PromptFlow silently returns zero rows because it runs as the agent user.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforcebuilding-voice-asa-agent
End-to-end playbook for building a voice Agentforce Service Agent (ASA) in a demo org — the full sequence from agent-user perm set, data seeding, system-context data flows, prompt template, agent bundle authoring, live-preview checkpoint, publish/activate, through inbound + per-channel escalation routing flows and the voice-channel Setup handoff. Ties together the system-context-flow, agent-user-perms, data-library, and escalation-flow skills into one ordered build.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforcecase-management-setup
Salesforce Case Management setup and troubleshooting. Helps admins and consultants configure cases, assignment rules, queues, Omni-Channel, Email-to-Case, escalations, entitlements, SLAs, Service Console, and sharing models. Diagnoses common setup issues.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforceconnecting-channels-to-asa
The connect-the-channels map for an Agentforce Service Agent (ASA): how a live VOICE call and an in-app/web MESSAGING chat get INTO the agent (inbound omni-flow → routeWork routingType=Copilot → BotDefinition) and how the agent hands a conversation OUT to a human — to a QUEUE, to a SKILL, or DIRECT to a specific rep — on @utils.escalate. Nails the two things people get wrong every time: (1) messaging escalation binds in the agent .agent metadata but VOICE binds only in Setup (never metadata); (2
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforcedeploying-person-account-fields-via-contact
Add custom fields to Person Accounts correctly by creating them on the CONTACT object — Salesforce auto-materializes each one on the Account object with a pc-suffixed API name (e.g. MemberStatusc on Contact → Account.MemberStatuspc), queryable via SOQL and the REST/Bulk API and referenceable in Flow. Explains why you must NOT create the field on Account directly for Person Accounts, how to read the mirrored field via SOQL/API, how roll-up summary fields are the one exception that must live on Ac
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.
sshvarzman-salesforcehandling-locked-standard-objects
Detect and work around standard industry/FSC objects (e.g. Claim, KeyPrefix 0Zk) that ACCEPT CustomField metadata deploys and Tooling API creates with success:true, report EntityDefinition.IsCustomizable=true, but NEVER materialize the physical column at runtime — the field is invisible to SOQL, REST describe, and Apex getGlobalDescribe. Confirm the trap with getGlobalDescribe, then use a standard long-text field with structured delimited lines as the workaround.
- Installations / 1h
- 1
- GitHub Stars
- 0
- Mis à jour
- 15 oct.