CtrlK
BlogDocsLog inGet started
Tessl Logo

cherry-tool-guide

Cherry Studio first-party tool and bundled-shell routing for general agents. For straightforward local work in shell-capable sessions, run JS/TS with `bun <file>` and one-off JS tools with `bun x`; run Python with `uv run [--with <pkg>] python` and one-off Python CLIs with `uvx`; search with `rg`. Load this guide before changing project dependencies, deciding whether a tool should be ephemeral or reusable, reading or converting local Office/PDF files, coordinating or delegating across Agent Sessions, or using Cherry-owned web/browser, knowledge, persistent memory, schedules/notifications, IM channels, image generation, artifact reporting, managed CLI, skill, or MCP-server-registration capabilities—even if the user names no tool. Consult it before shell/file workarounds; live tool schemas are authoritative.

76

Quality

96%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

SKILL.md
Quality
Evals
Security

Quality

Content

100%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

An exemplary router-style skill body: dense, policy-only content with concrete fully-qualified tool sequences, explicit availability/approval/error checkpoints, and a verified one-level-deep reference structure. It assumes Claude's competence and defers argument detail to authoritative live schemas by explicit design.

DimensionReasoningScore

Conciseness

"This file is a router. It carries only the global rules and the intent → tool → reference table" — every section is non-obvious policy (mutation boundaries, approval gating, error protocol) with zero padding and no explanation of concepts Claude already knows. Lean and efficient; every token earns its place.

5 / 5

Actionability

The routing table gives fully qualified tool names with concrete sequences ("`mcp__cherry-tools__kb_list` → `kb_search` → `kb_read`", "`command -v` check → `cli_list` → `cli_search` → `cli_install`", "run JS/TS with `bun <file>`"), and the body explicitly justifies not restating argument shapes ("the live tool schema in your session is the authoritative source"). Covers the common routing cases concretely; not a 4, which would require gaps in the executable guidance.

5 / 5

Workflow Clarity

Explicit checkpoints ("Check availability first", "resolve IDs first; needs approval") plus a feedback loop for errors ("read the message and correct the call; don't silently retry the same arguments", declined approval → "stop and report — never re-attempt the mutation through a different route") and a diagnostic section distinguishing absent tools from missing dependencies. Matches the explicit-validation-and-recovery anchor; mutation operations are gated, so the destructive-operation cap does not apply.

5 / 5

Progressive Disclosure

A clear overview with a routing table pointing to exactly one reference per intent; all ten linked files (web.md, knowledge.md, documents.md, memory.md, autonomy.md, sessions.md, outputs.md, cli.md, skills.md, mcp.md) exist in references/ and cross-link only laterally — one level deep, well-signaled, easy navigation. Matches the clear-overview-with-one-level-references anchor.

5 / 5

Total

20

/

20

Passed

Description

92%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description: explicit what and when, concrete run-command routing, comprehensive capability enumeration, and a well-bounded niche. The only weakness is trigger phrasing that leans on product-internal jargon over the natural terms a user would say, keeping keyword coverage just short of comprehensive.

Suggestions

Add a few natural user-voiced synonyms to the trigger list (e.g. "search the web", "remind me / scheduled tasks", "chat or IM channels") so the description fires on how users actually phrase these requests, not only on Cherry-internal terminology.

Trim internal jargon in the when-clause ("deciding whether a tool should be ephemeral or reusable", "bundled-shell routing") in favor of plainer equivalents ("choosing between a one-off command and an installed tool"), which also shortens the already-long description.

DimensionReasoningScore

Specificity

Lists multiple specific concrete actions with exact commands — "run JS/TS with `bun <file>`", "one-off JS tools with `bun x`", "run Python with `uv run [--with <pkg>] python`", "one-off Python CLIs with `uvx`", "search with `rg`" — plus comprehensive coverage of every routed capability domain (web/browser, knowledge, memory, schedules/notifications, IM, image generation, CLI, skills, MCP registration). Not a 4: no notable coverage gaps.

5 / 5

Completeness

Explicitly answers both: what — "Cherry Studio first-party tool and bundled-shell routing for general agents" — and when — "Load this guide before changing project dependencies, … reading or converting local Office/PDF files, … or using Cherry-owned … capabilities—even if the user names no tool." Concrete trigger phrases with an explicit load condition.

5 / 5

Trigger Term Quality

Good natural keywords ("dependencies", "Office/PDF files", "Agent Sessions", "schedules/notifications", "image generation") but coverage is not comprehensive: common user phrasings like "search the web", "remind me", or "chat channels" are absent, and several triggers are internal jargon ("Cherry-owned", "bundled-shell routing", "ephemeral or reusable"). Clearly above the some-relevant-keywords anchor (3) but below the synonyms-and-extensions anchor (5).

4 / 5

Distinctiveness Conflict Risk

A clear niche ("Cherry Studio first-party tool and bundled-shell routing") with distinct, product-specific triggers ("MCP-server-registration capabilities", "Cherry-owned web/browser"); minimal risk of firing for the wrong skill. Not a 4: it is not merely "mostly distinct" — the scope is unambiguously owned by this one product surface.

5 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
CherryHQ/cherry-studio
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.