sales-chatbot
Chatbot marketing and conversational automation — building chatbot flows, multichannel messaging (WhatsApp, Telegram, Facebook, Instagram, Viber, live chat), lead qualification bots, FAQ bots, and handoff to human agents. Use when website visitors leaving without engaging, WhatsApp or Telegram leads not getting qualified, bot flow feels broken or confusing, leads dropping off before reaching a human agent, or not sure how to automate conversations for sales. Do NOT use for live chat widget setup without bots (use /sales-live-chat), email sequences (use /sales-cadence), or SMS campaigns without conversational flow (use /sales-sms-marketing). For SendPulse-specific help, use /sales-sendpulse. For SWAI-specific help, use /sales-swai.
What this skill does
# Chatbot Marketing & Conversational Automation
Help the user build chatbot flows for sales, support, and marketing — across WhatsApp, Telegram, Facebook Messenger, Instagram, Viber, website live chat, and other messaging platforms.
## Step 1 — Gather context
If `references/learnings.md` exists, read it first for accumulated knowledge.
**What channel are you building for?**
- a) WhatsApp Business
- b) Telegram
- c) Facebook Messenger
- d) Instagram DMs
- e) Website live chat widget
- f) Viber
- g) Multiple channels / omnichannel
- h) Not sure yet — help me choose
**What's the bot's primary job?**
- a) Lead qualification — ask questions, score, route to sales
- b) FAQ / support — answer common questions automatically
- c) Product catalog / e-commerce — browse, recommend, purchase
- d) Appointment booking — schedule meetings or demos
- e) Welcome / onboarding — greet new subscribers, set expectations
- f) Re-engagement — win back inactive contacts
- g) Order tracking / transactional — status updates, notifications
- h) Full conversational sales funnel
**What's your technical comfort level?**
- a) No-code visual builder only
- b) Comfortable with webhooks and basic integrations
- c) Developer — want API-level control
## Step 2 — Strategy and approach
**Read `references/platform-guide.md`** for channel comparison, flow design principles, lead qualification frameworks, FAQ bot structure, platform-specific guidance (SendPulse, SWAI, ManyChat, Tidio, Intercom, Drift, HubSpot), bot-building steps, metrics benchmarks, and common mistakes.
*You no longer need the platform guide details — focus on the user's specific situation.*
## Step 3 — Actionable guidance
Based on the user's channel and bot type, provide:
1. **Platform recommendation** — which chatbot platform fits their channel, budget, and technical level
2. **Flow design** — map the conversation for their specific use case (qualification, FAQ, e-commerce, booking)
3. **Integration setup** — connect bot to CRM, calendar, or e-commerce platform
4. **Testing plan** — how to test the bot before going live
5. **Optimization** — which metrics to track and how to iterate on drop-off points
## Gotchas
*Best-effort from research — review these, especially items about platform-specific rules that may change.*
- **WhatsApp 24-hour window is strict** — if you send a non-template message outside the window, it silently fails or gets your account flagged. Always check `last_activity` timestamp before sending.
- **Facebook/Instagram 24-hour rule** — similar to WhatsApp, Messenger has a 24-hour standard messaging window. Outside it, you need message tags (limited to specific use cases like confirmed event updates or account updates).
- **Template approval takes time** — WhatsApp message templates require Meta review (hours to days). Don't wait until launch day to submit templates.
- **Bot-to-human handoff state** — most platforms pause the bot during human chat but don't always resume correctly. Test the full cycle: bot → human → bot resumes.
- **Multi-channel subscriber identity** — a user on WhatsApp and Telegram is often two separate contacts in the platform. Merging/deduplication requires matching by phone or email.
- **Self-improving**: If you discover something not covered here, append it to `references/learnings.md` with today's date.
## Before recommending a specific platform skill
This skill covers a strategy domain across many platforms. **Before pointing the user to any specific platform skill** (any `/sales-{platform}` listed in `## Related skills`, e.g., `/sales-mailshake`, `/sales-klaviyo`, `/sales-apollo`), read that platform skill's actual `SKILL.md` first. The 1-line description in `## Related skills` is enough to *identify* a candidate — it's not enough to *commit* to it or to write a prompt that invokes it well.
**How to read it:**
- If `~/.claude/skills/{skill-name}/SKILL.md` exists locally, `Read` it.
- For `sales-*` skills, `WebFetch` directly from this repo: `https://raw.githubusercontent.com/sales-skills/sales/main/skills/{skill-name}/SKILL.md` — e.g., for `sales-mailshake`: `https://raw.githubusercontent.com/sales-skills/sales/main/skills/sales-mailshake/SKILL.md`.
- For non-`sales-*` skills (third-party), look up `{org}/{repo}` in `~/.claude/skills/sales-do/references/skill-sources.md` if installed and fetch the same `skills/{skill-name}/SKILL.md` path under that repo.
**After reading,** ground your recommendation in something concrete from the SKILL.md (its scope, a sub-flow, its `argument-hint` shape, or a "Do NOT use for..." negative trigger). Align any generated invocation with the platform skill's `argument-hint`. If the platform skill turns out not to fit the user's situation, swap to another or handle the question here directly rather than recommending a poor fit.
## Step 5 — Related skills
- `/sales-zendesk` — Zendesk platform help (AI agents, chatbot automation, ticketing)
- `/sales-helpdesk-selection` — Help desk platform comparison (Zendesk vs Freshdesk vs Intercom vs Help Scout)
- `/sales-hubspot` — HubSpot platform help (chatbot builder, Conversations inbox, live chat)
- `/sales-qualified` — Qualified platform help (Piper AI SDR conversations, Salesforce-native chatbot)
- `/sales-sendpulse` — SendPulse platform help (chatbots across 7 channels, visual flow builder, Automation 360)
- `/sales-swai` — SWAI platform help (AI chat widget, autonomous lead qualification, Cal.com scheduling)
- `/sales-birdeye` — Birdeye platform help (webchat chatbot for multi-location businesses, appointment scheduling)
- `/sales-live-chat` — Live chat widget setup and optimization (non-bot focused)
- `/sales-sms-marketing` — SMS campaigns and automation (non-conversational)
- `/sales-cadence` — Multi-channel outbound sequences (email, call, social — structured outreach vs. conversational)
- `/sales-email-marketing` — Opt-in email marketing (broadcasts, automations, newsletters)
- `/sales-integration` — Connect chatbot platforms to CRM, helpdesk, and other tools
- `/sales-audience-growth` — Grow your subscriber base (pop-ups, lead magnets, opt-in forms — feed contacts into bot flows)
- `/sales-lindy` — Lindy platform help (AI agent builder with custom chatbot creation and knowledge base RAG)
- `/sales-do` — Not sure which skill to use? The router matches any sales objective to the right skill. Install: `npx skills add sales-skills/sales --skill sales-do`
## Examples
### Example 1: WhatsApp lead qualification bot
**User says**: "I want to build a WhatsApp bot that qualifies leads for my SaaS product and books demos"
**Skill does**: Recommends WhatsApp Business API setup, designs a 4-step qualification flow (greeting → role question → team size → pain point), explains 24-hour window and template requirements, configures meeting booking via calendar link, sets up hot lead handoff to sales rep
**Result**: Working WhatsApp bot that qualifies inbound leads and books demos for qualified prospects
### Example 2: Multi-channel FAQ bot
**User says**: "We get the same 20 questions on our website, Facebook, and Instagram. Can I automate this?"
**Skill does**: Recommends a platform with multi-channel support (SendPulse, ManyChat, or Tidio), structures FAQ into 5 categories with sub-menus, configures fallback to human agent after 2 failed attempts, sets up the same flow across website widget + Facebook + Instagram
**Result**: FAQ bot deployed on 3 channels, reducing support ticket volume by handling common questions automatically
### Example 3: E-commerce abandoned cart recovery via Messenger
**User says**: "I want to send abandoned cart reminders through Facebook Messenger instead of email"
**Skill does**: Explains Messenger 24-hour window and checkbox plugin for opt-in, designs flow: 1hr reminder with product image → 24hr discount offer → fallback to email if no response, configures Shopify/WooCommerce webhook to trigger the flow, adds human handoff for purchase objections
**Result**: MeRelated 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.