Chatbot conversation design — persona, intent mapping, slot filling, error recovery
Design resilient chatbot conversations — persona and scope, intent/entity mapping, slot-filling and disambiguation dialogs, three-tier fallback, context carryover, testing
- What
- Design resilient chatbot conversations — persona and scope, intent/entity mapping, slot-filling and disambiguation dialogs, three-tier fallback, context carryover, testing
- Cost
- Free
- Needs
- a chatbot project to design (new or existing) — the bot's purpose, target users, and the list of tasks it should handle
- Install
- Copy the installer prompt below into your Muse — your agent does the rest.
Curated by Skill Harbor — @seb1n's chatbot-conversation-design skill: a full-lifecycle framework for designing chatbot conversations that survive real users. The agent defines the bot persona and scope first (personality, guardrails, a style guide — a well-scoped bot that nails five tasks beats a vague one attempting fifty), maps intents and entities with journey graphs covering happy paths, branches, and exits, designs slot-filling sequences and disambiguation dialogs (with confirmation prompts for high-stakes actions), builds a three-tier error recovery system (rephrase, constrained options, human escalation — never a dead end), implements context carryover and dialog-state stacks for mid-conversation topic switches, then prototypes, tests, and iterates against metrics like task completion rate and fallback trigger rate. It ships two worked examples (a complete customer-support order-tracking flow and a fallback/escalation flow with state notes), best practices (design the unhappy path first, keep bot utterances under three sentences, confirm irreversible actions, progressive disclosure), and edge-case guidance (multi-intent inputs, topic switching, contradictory info, emotional users, platform constraints like WhatsApp limits). Honest caveats: a design methodology, not a working bot — you still build the bot yourself; the prompt-format template expects you to supply the use case, persona, tasks, and platform. MIT licensed (frontmatter and manifest agree). Skill Harbor never reviews the code, review it yourself before use. Discovered via skills.sh.
Version:
Install
Prerequisites: a chatbot project to design (new or existing) — the bot's purpose, target users, and the list of tasks it should handle Install "Chatbot conversation design — persona, intent mapping, slot filling, error recovery" for me. It gives my agent @seb1n's conversation-design framework: define the bot persona, scope, and style guide; map every intent and entity with journey graphs (happy paths, branches, exits); design slot-filling sequences and disambiguation dialogs with confirmations for high-stakes actions; build three-tier fallback (rephrase, constrained options, human escalation) so no conversation dead-ends; implement context carryover and dialog-state stacks for topic switches; then prototype, test against completion/fallback metrics, and iterate — following the worked order-tracking and error-handling examples, the best practices (unhappy path first, short bot utterances, progressive disclosure), and the edge-case guidance (multi-intent, contradictory info, emotional users, platform constraints). MIT licensed. Repository: https://github.com/seb1n/awesome-ai-agent-skills/blob/main/communication/chatbot-conversation-design/SKILL.md 1. Fetch the SKILL.md file (and any helper files) from the repository path into a temporary folder and summarize what it does in one or two sentences. 2. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code. Verify that holds here; STOP on any red flag and tell me. 3. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "chatbot-conversation-design". 4. Verify with no network calls: frontmatter valid, files in place. 5. Report what was installed, where, and what I still need to do myself (e.g. give the agent the bot's use case, persona, supported tasks, and target platform so it can design the flows). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.
Questions
How do I install a build?
Every product page includes a copy-paste install prompt. Paste it into your Muse and it sets the build up for you — no manual configuration.
Where does my money go?
Straight to the seller. Skill Harbor never processes payments: checkout happens on the seller’s own page, usually Stripe.
What does the ✓ next to a creator’s name mean?
It means we confirmed the identity of the person behind the listing. It says nothing about the code itself — always check a build before installing it.