Skip to content
skillxone

CherryHQ/cherry-studio / resources/skills/cherry-tool-guide

cherry-tool-guide

Verified

Cherry tool routing: bun, uv, rg, knowledge, memory, schedules, skills, MCP, web. Read before shell workarounds.

Cherry Tool Guide

Cherry Studio provides first-party tools through four MCP servers: cherry-tools, agent-memory, skills, and mcp-manager. Which tools mount depends on the session. Shell-capable general agents also get 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.

Resolve server/tool identity to a callable name: take the exact identifier verbatim from your current session's schema or discovery result. Treat every qualified name in this guide and its references as a runtime-specific example governed by this rule. Never construct a name from examples, convert punctuation, convert to camelCase, or reconstruct a digest. 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

  • Check availability first. Apply this procedure to every absence claim throughout this guide and its references, including claims based on a missing live-list entry. Missing from initial tool declarations does not mean absent: a tool may already be active or reachable through deferred/script discovery. First inspect tools currently available, then use whatever discovery your session actually offers. Query by capability or bare tool/server identity (e.g. memory, agent-memory), not only an assumed full identifier. On a miss, broaden the query or inspect the catalog as the discovery tool's own schema allows. Only after these checks establish absence, stop and report the capability absent. If discovery is unavailable or fails, report that limitation instead of asserting nonexistence. Never pretend a call succeeded or fabricate a result.
  • Don't reach around Cherry's mutation boundaries. Knowledge bases, IM channels, 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.
  • Honor approval. mcp__cherry-tools__kb_manage, mcp__cherry-tools__cli_install, mcp__cherry-tools__session_create, mcp__cherry-tools__session_send, mcp__skills__install_skill, and mcp__mcp-manager__install_mcp_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.
  • Intent still gates auto-approved effects. Memory writes, schedule changes, notifications, Session renames, 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.

Routing table

User intentRoute toReference
Look up current/online facts, news, docsmcp__cherry-tools__web_search → mcp__cherry-tools__web_fetchweb.md
Browser interaction (click, forms, screenshots)Automation REPL when Browser is enabled; load cherry-browserweb.md
Native desktop applications (open, type, click, screenshots)Automation REPL when Computer Use is enabled; load cherry-computerautomation.md
Answer from the user's own documentsmcp__cherry-tools__kb_list → mcp__cherry-tools__kb_search → mcp__cherry-tools__kb_readknowledge.md
Add / delete / re-index knowledgemcp__cherry-tools__kb_manage (resolve IDs first; needs approval)knowledge.md
Read or convert a local documentmcp__cherry-tools__to_markdown → read the returned temporary Markdown path as neededdocuments.md
Recall a past fact, correction, or preferenceServer agent-memory, tool memory (search) before re-asking; resolve callable name abovememory.md
Save durable knowledge vs. a one-off eventServer agent-memory, tool memory (update vs. append); resolve callable name abovememory.md
Schedule a recurring / future taskmcp__cherry-tools__cron (Cherry scheduling only)autonomy.md
Proactively message the user or send a filemcp__cherry-tools__notify (needs configured recipients)autonomy.md
Inspect / connect / repair IM channels, rename agentmcp__cherry-tools__configautonomy.md
Rename the current or another Agent Sessionmcp__cherry-tools__session_renamesessions.md
Find, create, message, or inspect work across Agent Sessionsmcp__cherry-tools__session_list / session_search / session_create / session_send / session_deliveriessessions.md
Generate an imagemcp__cherry-tools__generate_image (needs a painting model)outputs.md
Declare final deliverable file(s)mcp__cherry-tools__report_artifactsoutputs.md
Run JS/TS or Python, invoke a one-off package, search local code/filesbundled bun, uv / uvx, or rg according to task lifetimecli.md
Find / install a command-line toolcommand -v check → mcp__cherry-tools__cli_list → mcp__cherry-tools__cli_search → mcp__cherry-tools__cli_install (approval)cli.md
Find / install a new capability skillmcp__skills__search_skills → mcp__skills__install_skill (approval)skills.md
Register a new MCP server the user suppliedmcp__mcp-manager__install_mcp_server (approval; never invent the config)mcp.md

When a tool isn't there

Distinguish these cases after checking availability as above:

  • The tool exists but is deferred or already loaded. It may be missing from initial declarations, and a loading search may return nothing. That means no new matching tools were loaded, not that the tool does not exist. Re-check tools already available and any earlier successful discovery; use script/catalog discovery when your session offers it. An empty catalog query means no matches in that query's searched set; neither result alone proves absence.
  • The capability is unavailable after resolution. For example, no knowledge base is in scope, or CLI management is disabled for a shell-less agent. Explain what's missing and what the user can configure; don't work around it with shell/file tools. Use the domain reference's availability conditions under the shared procedure above.
  • The tool is listed but reports a missing dependency → e.g. mcp__cherry-tools__generate_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.

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 read_file attachment reader (a chat-path tool, not exposed on this MCP surface); and the role-specific mcp__assistant__* navigation/diagnosis tools, which belong to the Cherry Assistant and its own guide.

reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
discernment-nudgeanthropics/skillsAfter you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.Stars: 195kUpdated Claude CodeCodexVerifiedInstall
internal-commsanthropics/skillsA set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).Stars: 195kUpdated Claude CodeCodexVerifiedInstall