kin
How to use Kin's MCP tools (search, describe, read, execute, feedback) and read their answers. Use when the person asks to read or change something in a service they connected in Kin, such as GitHub, Notion or Linear.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-14
- Stars
- ★ 1.6k
- Forks
- 491
01 前言 Front Matter
- name
- kin
- description
- How to use Kin's MCP tools (search, describe, read, execute, feedback) and read their answers. Use when the person asks to read or change something in a service they connected in Kin, such as GitHub, Notion or Linear.
03 SKILL.md 正文
Using Kin
Kin connects the person's accounts at other services. For each account and action the person sets a rule: Allow, Ask me or Block. Kin applies the rule on its server when the action runs, so these notes only cover how to make the calls and read the answers.
Find the action
- Call
searchwith the task in plain words, such as "list open issues in GitHub". It returns only the actions the person's rules allow, and each match names the tool that runs it:readorexecute. - A match that includes
inputSchemacan run right away. Calldescribe(up to 5actionIdsat once) when you need an action's input schema, its current version, or whether it needs the person's approval. noMatchmeans nothing the person connected clearly does the task. Tell them.suggestionsare services they could connect; mention them, and connect one only if the person asks.
Run it
- Run reads with
read. - Run changes with
execute. Give each new change a newrequestKeyand a one-sentencereasonthat says why, in your own words. The person sees the reason when they review the call and in Kin's activity log. - The same
requestKeynever runs a change twice. Reuse a key only to get the outcome of the change it started.
Read the answer
approval_required: the change is waiting for the person in Kin. Tell them, and include thereviewUrl. Once they decide, make the same call with the samerequestKeyto get the outcome.running: the person approved it and it is still running. Make the same call again to check on it.denied: the person's rule does not allow this. Tell them what was refused.error.messagesays why.paused: the person paused AI access in Kin. Stop and tell them.unknown, ortimed_outfor a change: it may have happened. Check before you try again with a new key.trimmed: the result was too long and was cut.trimmed.nextreads the rest.ambiguity: more than one connected account fits. See thekin-accountsskill.
Feedback
Call feedback only when a search match or a call was wrong, or notably useful. It tunes search ranking and never changes permissions. Routine successes need none.
04 同仓库更多 Skills
worktable-review-with-annotations
Create, inspect, reply to, update, or resolve situated Worktable annotations. Use when feedback belongs to a Doc section, block, text range, or HTML Doc, or when the user asks to address existing review feedback and coordinate a requested artifact edit.
worktable-find-and-synthesize-context
Find, assess, and synthesize context from Worktable Docs and Records. Use when the user asks what a workspace contains, wants workspace-backed research, needs current facts reconciled across artifacts, or requests a cited synthesis without changing workspace state.
worktable-create-or-update-docs
Create, edit, rename, or maintain narrative Worktable Docs. Use for plans, decisions, briefs, notes, research syntheses, and other durable prose when the user requests a new artifact or a safe update to an existing one.
worktable-create-or-manage-records
Organize repeated structured information as durable Worktable Records. Use for research datasets, inventories, directories, pipelines, trackers, and other collections whose items need identity, fields, filtering, links, validation, or individual updates.
worktable-create-or-manage-interactive-html
Create or improve durable visual and interactive artifacts in Worktable. Use for dashboards, trackers, calculators, mini apps, slide presentations, and other requests that benefit from a polished HTML Doc.
worktable-collaborate-threads
Read, answer, and hand off messages through durable Worktable threads with users or other authenticated participants. Use when the user asks to check addressed messages, reply to a participant, continue a thread, request another agent's help, wait briefly for a response, or watch for comments and messages sent to the agent.