cherry-tool-guide skill

Cherry Tool Guide

Cherry Studio injects first-party tools into your session over four MCP servers (mcpcherry-tools, mcpagent-memory, mcpskills, mcpmcp-manager) and gives shell-capable general agents bundled runtimes for local execution. The MCP tools act on the running app — the user's knowledge bases, IM channels, schedules, managed CLIs, skill library, and MCP server registry — through boundaries only Cherry owns. Shell and file tools cannot reach those app boundaries correctly; use the bundled runtimes only for the local execution cases routed below.

This file is a router. It carries only the global rules and the intent → tool → reference table. Each reference holds that domain's prerequisites, sequencing, conditional availability, output interpretation, recovery, and examples. Read the one reference the task needs (and any it cross-links to) before calling — don't work from this page alone.

Tool names here are fully qualified (mcpservertool); the exact names exposed in your session are authoritative if they ever differ. This guide never restates argument shapes — the live tool schema in your session is the authoritative source for parameter names, enums, and required fields. Read it before every call.

Global rules

which). If a tool is not in your live tool list, its capability is unavailable in this session — say so honestly and stop; never pretend a call succeeded or fabricate a result.

• Check availability first. Several tools are conditional (each reference says

schedules, managed CLIs, skills, and registered MCP servers are mutated only through these tools. Do not shell out to npm install, git clone, crontab, or hand-edit knowledge or MCP settings files to accomplish these — the tool does bookkeeping (registration, scoping, approval, sync) that a raw shell command skips. Shell is fine for inspection (e.g. command -v to probe PATH) — just not to perform the owned mutation.

• Don't reach around Cherry's mutation boundaries. Knowledge bases, IM channels,

mcpcherry-toolssessioncreate, mcpcherry-toolssessionsend, mcpskillsinstallskill, and mcpmcp-managerinstallmcp_server are gated by the session's approval mode. Call them only once the user's intent is clear; if approval is declined, stop and report — do not retry the same effect through another route.

• Honor approval. mcpcherry-toolskbmanage, mcpcherry-toolscliinstall,

notifications, and agent/channel configuration may execute without an approval card. Do not call them merely because they are available; first make sure the user requested the effect or it is necessary to complete an already-approved task.

• Intent still gates auto-approved effects. Memory writes, schedule changes,

Routing table

When a tool isn't there

Two different situations, don't confuse them:

session (e.g. no knowledge base in scope, or CLI management disabled for a shell-less agent). Explain what's missing and what the user can do; don't work around it with shell/file tools. The reference for that domain says exactly when it can be absent.

• The tool is absent from your live list → the capability is unavailable this

mcpcherry-toolsgenerate_image with no painting model. It stays listed and returns a note; relay the note and point the user at configuration — don't retry blindly or fake success.

• The tool is listed but reports a missing dependency → e.g.

On any tool error result (bad ID, unsupported channel/file, invalid recipe), read the message and correct the call; don't silently retry the same arguments. On declined approval, stop and report — never re-attempt the mutation through a different route.

Out of scope

Not covered here: SDK-native Read/Edit/Bash and orchestration tools; calling the tools of a third-party (user-configured) MCP server — only registering one is in scope, see mcp.md; the AI-SDK chat readfile attachment reader (a chat-path tool, not exposed on this MCP surface); and the role-specific mcpassistant__* navigation/diagnosis tools, which belong to the Cherry Assistant and its own guide.