lindy-core-workflow-a
Build and configure multi-step Lindy AI agent workflows. Use when creating agents with triggers, actions, conditions, knowledge bases, or agent steps. Trigger with phrases like "create lindy agent", "build lindy agent", "lindy agent workflow", "configure lindy agent", "lindy workflow".
What this skill does
# Lindy Core Workflow A: Agent Creation ## Overview Complete workflow for creating, configuring, and testing Lindy AI agents. Agents consist of four components: **Prompt** (behavioral instructions), **Model** (AI engine), **Skills** (available actions), and **Exit Conditions** (completion criteria). ## Prerequisites - Lindy account at - Use case defined (support bot, email triage, data processor, etc.) - Required integrations identified (Slack, Gmail, Sheets, etc.) ## Instructions ### Step 1: Define Agent Specification Before building, document: - **Purpose**: What the agent does (one sentence) - **Trigger**: What wakes it up (webhook, email, schedule, chat, etc.) - **Actions**: What it does (send email, update sheet, post to Slack, etc.) - **Model**: GPT-4 (smart, expensive), Claude Sonnet (balanced), Gemini Flash (fast, cheap) ### Step 2: Create the Agent **Option A — Natural Language (recommended)**: 1. Click **New Agent** at 2. Describe your agent in plain English: ``` When a customer emails [email protected], classify the email as billing/technical/general, draft a response using our knowledge base, and post the classification to #support-triage in Slack ``` 3. Agent Builder auto-generates trigger + action nodes **Option B — Manual Build**: 1. Click **New Agent** > **Start from scratch** 2. Add trigger: Click **"+"** > Select trigger type 3. Add actions: Click **"+"** > Search for action 4. Connect nodes by dragging arrows between steps ### Step 3: Configure the Prompt Open **Settings > Prompt**. Structure it with clear sections: ``` ## Identity You are a customer support classifier and responder for [Company]. ## Instructions 1. Read the incoming email carefully 2. Classify into: billing, technical, or general 3. Search the knowledge base for relevant answers 4. Draft a professional response using the KB results 5. If no KB match found, escalate to human ## Constraints - Never promise refunds or credits without human approval - Keep responses under 200 words - Always include ticket reference number ``` **Prompt best practices** (from Lindy docs): - Use action-oriented language (imperatives) - Break complex logic into numbered steps - Include few-shot examples for consistent formatting - Add constraints to prevent unwanted behavior ### Step 4: Add Conditions (Branching Logic) 1. Click **"+"** > **Condition** 2. Write natural language condition: `"Go down this path if the email is about billing"` 3. Add multiple branches for different classifications 4. Enable **"Force the agent to select a branch"** for deterministic routing ### Step 5: Configure Actions For each action, set field modes: **Auto mode** — Agent infers the value from all previous step data: ``` Best for: predictable mappings where field names align ``` **AI Prompt mode** — Give natural language instructions: ``` Summarize the email in 2 sentences, then include the classification. Reference: {{email_received.body}} ``` **Set Manually mode** — Exact value, no AI: ``` Channel: #support-triage ``` ### Step 6: Add Knowledge Base (Optional) 1. Go to **Settings > Knowledge Base** 2. Add sources: PDF, DOCX, Google Drive, Notion, websites 3. Configure search: - **Max Results**: 4 (default) to 10 - **Search Fuzziness**: 0 (keyword) to 100 (semantic, recommended) 4. Agent auto-searches KB when relevant to the task ### Step 7: Test the Agent 1. Use the **Test** button in the workflow editor 2. Provide sample input matching your trigger type 3. Review each step's output in the task detail view 4. Iterate on prompt and action configuration ## Trigger Types Reference | Trigger | Use Case | Configuration | |---------|----------|---------------| | **Webhook Received** | External API calls | URL + secret key | | **Email Received** | Inbox automation | Gmail/Outlook + label filters | | **Schedule** | Recurring tasks | Cron-style: daily, weekly, custom | | **Chat Message** | Interactive bot | Lindy Chat or Embed widget | | **Slack Message** | Team automation | Channel + keyword filters | | **Agent Message** | Multi-agent delegation | Receives from other Lindies | | **Calendar Event** | Meeting automation | Minutes offset (-30 = 30 min before) | | **Form Submission** | Lead capture | Connected form integration | ## Action Categories | Category | Actions | |----------|---------| | **Email** | Send Email, Draft Reply, Search Inbox, Add Label | | **Slack** | Send Channel Message, Send DM, Thread Reply | | **Sheets** | Update Spreadsheet, Get Document | | **Calendar** | Create Event, Reschedule, Cancel | | **Knowledge** | Search Knowledge Base, Resync KB | | **Code** | Run Code (Python/JS in E2B sandbox) | | **Web** | HTTP Request, Web Search, Website Crawler | | **Memory** | Read/Create/Update/Delete Memory | | **Phone** | Make Call, Transfer Call, End Call | | **Agent** | Agent Send Message (delegation) | ## Error Handling | Error | Cause | Solution | |-------|-------|----------| | "No trigger configured" | Agent has no trigger | Add at least one trigger node | | Action fails silently | Wrong field mode | Switch to AI Prompt or Set Manually | | KB returns no results | Fuzziness too low | Increase to 100 (semantic search) | | Condition always picks same path | Ambiguous prompt | Make conditions more specific | | Agent loops indefinitely | No exit condition | Add measurable exit criteria | ## Resources - [Lindy Introduction](https://docs.lindy.ai/fundamentals/lindy-101/introduction) - [Triggers Documentation](https://docs.lindy.ai/fundamentals/lindy-101/triggers) - [Actions Documentation](https://docs.lindy.ai/fundamentals/lindy-101/actions) - [Prompt Guide](https://docs.lindy.ai/fundamentals/lindy-101/prompt-guide) ## Next Steps Proceed to `lindy-core-workflow-b` for triggers, automation, and multi-agent delegation.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.