Conversation state human handoff
Skill gaelic-ghost/socket/skills/conversation-state-human-handoff
Plan conversation identity, state, consent, escalation, transcript, and human-handoff behavior for bots and business messaging agents.From its SKILL.md
npx -y skills add gaelic-ghost/socket --skill conversation-state-human-handoffAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 6 stars6 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
1.6 KB, 268 tokens by cl100k_base, as published. Nobody here has run it
Conversation State And Human Handoff
Purpose
Design a narrow, explainable conversation boundary. Platforms identify people, installations, channels, and conversations differently, so normalize only the application facts needed for the product; do not erase platform identity or policy distinctions.
Workflow
- Identify the platform-native sender, conversation, installation or tenant, message, and consent identifiers.
- State which identifiers are durable, which can change, and which are sensitive.
- Define the minimal state required for the user experience: task progress, user preference, handoff status, and delivery context.
- Specify an explicit escalation path: trigger, acknowledgement, operator queue or destination, visible transition, return-to-automation rule, and unavailable-hours behavior.
- Apply platform and product consent/retention rules before storing transcripts, media, model inputs, or contact data.
- Test interruption, duplicate delivery, human takeover, failed escalation, opt-out, deletion, and resumed conversation behavior.
Guards
- Do not map identities across platforms as though they represent the same authenticated person without an explicit account-linking flow.
- Do not let an AI response conceal that a human has joined or left when the platform has a representative-state mechanism.
- Do not introduce repositories, managers, or state mirrors into an app without a concrete ownership need.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.
Gives 0 of the 12 instructions most agent orchestration skills give in 268 tokens
Counted across 848 of the 1,300 authors here whose files we hold, read 2026-09-06
- Dispatch one agent per independent problem domainin 56 of 848, across 42 files
- Run full test suite after integrationin 55 of 848, across 42 files
- Verify fixes do not conflictin 40 of 848, across 32 files
- Review each summary when agents returnin 40 of 848, across 31 files
- Write a handoff document summarising the current conversationin 30 of 848, across 25 files
- Reference existing artifacts by path or URLin 26 of 848, across 24 files
- Give each agent a specific scopein 19 of 848, across 10 files
- Give each agent a clear goalin 19 of 848, across 10 files
- Include a suggested skills section in the documentin 18 of 848, across 16 files
- Tailor the doc to the user argumentsin 18 of 848, across 15 files
- Issue all subagent dispatches in the same responsein 17 of 848, across 11 files
- Use git worktrees for isolationin 17 of 848, across 8 files
Said here and by no other author read
- Identify platform-native sender and conversation identifiers
- Define minimal state required for user experience
- Specify an explicit escalation path
- Apply platform and product consent rules
- Test interruption and human takeover behavior
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.