Read and edit shared mind maps from Claude, Cursor or any MCP client. Your AI assistant can search a map, add and rewrite nodes, link ideas and tidy the layout. Your team sees each change live on the canvas and can undo any edit from the map's history.
Scored across 4 tools
The four tools have clearly distinct roles: search_tools finds names, describe_tools returns schemas, run_read executes read-only operations, and run_write executes mutations. The read/write and search/describe splits are explicitly described, leaving no realistic chance of misselection.
All names use snake_case with a verb_noun pattern (search_tools, describe_tools, run_read, run_write), which is consistent and predictable. The only slight variance is run_read/run_write using 'run' as the verb, but this still fits the same convention.
Four tools is exactly the right size for a gateway pattern: discovery, introspection, and read/write execution cover the necessary workflow without bloat. Each tool has a clear, non-redundant purpose.
The surface covers the full gateway lifecycle (search → describe → execute) and supports both read and write operations, so agents can reach any underlying mind-map tool. A minor gap is that no explicit domain operations (e.g., list_maps, create_map) are exposed directly, forcing discovery, but the generic runners make the surface functionally complete.
Get the argument schema and an example call for Shiko tools you know by name. Call it before run_read or run_write; use search_tools to find a name.
| Name | Required | Description | Default |
|---|---|---|---|
| names | Yes | Up to 10 tool names, such as ["read_map", "edit_map"] |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With readOnlyHint=true and openWorldHint=false already declaring the safety profile, the description adds the key return semantics: the agent gets an argument schema plus an example call, which is essential given there is no output schema. It stops short of covering edge behavior, such as what happens for an unknown name, so it is not a full 5.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences with zero filler; the capability statement is front-loaded and the routing instructions follow. Every clause earns its place by changing agent behavior.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a one-parameter lookup tool with no output schema, the description covers the return payload and the call sequence, which is the main risk of misuse. Missing only failure-mode context for unknown names, a minor gap for this complexity level.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% and the single `names` parameter is documented with an example in the schema itself. The description only reinforces this with "tools you know by name," adding no format or batching guidance beyond the structured field, so the baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb and resource: retrieve the argument schema and an example call for named Shiko tools. It is immediately distinguishable from siblings, since search_tools finds names and run_read/run_write execute them, and this tool only describes.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
"Call it before run_read or run_write" gives an explicit ordering constraint, and "use search_tools to find a name" names the alternative and the condition that selects it. Nothing about when to reach for this tool versus a sibling is left to inference.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Run a Shiko tool marked [read], such as read_map. Never changes anything. Find the name with search_tools and the arguments with describe_tools first. An unknown name, a [write] tool or invalid arguments returns an error saying what to fix.
| Name | Required | Description | Default |
|---|---|---|---|
| args | No | That tool's arguments, in the shape describe_tools shows. Omit if it takes none. | |
| name | Yes | Tool name, such as read_map, from search_tools |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and openWorldHint=false, so the safety profile is covered; the description still adds that it "never changes anything" and, more usefully, discloses the failure mode: unknown names, write tools, and invalid arguments return an error saying what to fix. It stops short of describing what a successful call returns or whether underlying tool output is passed through verbatim.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Four short sentences, front-loaded with the core action and followed by the prerequisite workflow and error behavior. No filler or hedging; every sentence carries distinct information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a nested-argument dispatcher with no output schema, the description covers invocation, prerequisites, and failure modes well. The remaining gap is that it never says what a successful call returns (presumably the wrapped tool's result), which an agent would have to infer.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% and the schema already says name comes from search_tools and args follow the shape describe_tools shows, so the description mostly restates that. The one added nuance is that args may be omitted when the target takes none, which the schema only implies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
States a specific verb and resource ("Run a Shiko tool marked [read]") and gives a concrete example (read_map), so an agent immediately knows this is the dispatcher for read-only tools. It also implicitly distinguishes itself from run_write by naming [write] tools as an error case.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly prescribes the workflow: find the name with search_tools and the arguments with describe_tools first, then invoke. It also states the exclusion condition (a [write] tool is rejected), which is textbook when/when-not guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Run a Shiko tool marked [write], such as edit_map. Changes the user's maps: each call is atomic, and the user can undo it from the map's History panel. Find the name with search_tools and the arguments with describe_tools, and read guide {topic:"editing"} before your first write. An unknown name, a [read] tool or invalid arguments returns an error saying what to fix.
| Name | Required | Description | Default |
|---|---|---|---|
| args | No | That tool's arguments, in the shape describe_tools shows. Omit if it takes none. | |
| name | Yes | Tool name, such as read_map, from search_tools |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare destructiveHint=true and non-idempotent, so the safety profile is known; the description adds genuinely new behavior — each call is atomic and undoable from the map's History panel. It also documents the failure mode (unknown name, [read] tool, or invalid args returns an error explaining what to fix).
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Three short sentences, front-loaded with what the tool is and what it changes, then prerequisites, then error behavior. No sentence is redundant.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a generic write dispatcher with no output schema and nested free-form args, the description covers effect, atomicity, reversibility, prerequisites, and error semantics. Nothing an agent needs before invoking it is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and both parameters are documented in the schema, so the baseline is 3. The description earns a bump by telling the agent where the values come from in practice (search_tools for name, describe_tools for the args shape) and by noting args can be omitted when the target takes none.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description names a specific verb and resource ("Run a Shiko tool marked [write]") and gives a concrete example (edit_map), with the effect stated plainly ("Changes the user's maps"). This cleanly distinguishes it from the sibling run_read.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
It spells out the full workflow: get the name from search_tools, the arguments from describe_tools, and read guide {topic:"editing"} before the first write. It also implicitly routes read operations to the [read] sibling by stating a [read] tool returns an error here.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Find Shiko's map tools when you don't know the name you need. Returns one line per tool: name [read|write] summary. Next, call describe_tools for its arguments.
| Name | Required | Description | Default |
|---|---|---|---|
| query | No | Words to look for, such as "comments". Without one, or with no match, every tool is listed. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations cover the safety profile (readOnlyHint=true, openWorldHint=false), but the description adds real value by disclosing the compact return format line-by-line and the read/write tagging of each result. There is no output schema, so this preview is doing genuine work. It stops short of stating result limits or how large a full listing can get.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two short sentences, both front-loaded with the most decision-relevant information: the use case first, then the return shape and the next call. No filler and nothing an agent must read twice.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a one-parameter read-only discovery tool with no output schema, the description supplies the missing piece (what a result looks like) and the chaining rule. Only minor gaps remain, such as whether the full listing is paginated or capped.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% and the single query parameter is fully documented there, including the no-query/no-match behavior of listing everything. The description adds nothing about the query beyond restating the discovery intent, so the baseline 3 applies.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Names a specific verb (find) and resource (Shiko's map tools), scopes it to the discovery case ("when you don't know the name you need"), and previews the exact return shape (name [read|write] summary). It also routes to describe_tools as the follow-up, so an agent can place it in the search→describe→run sequence without opening a schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Gives an explicit when-to-use condition (you don't know the tool name), which implicitly excludes the case where the name is known, and names the next tool to call for arguments. The workflow handoff to describe_tools is spelled out rather than left to inference.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
describe_toolsrun_readrun_writesearch_tools
Add one secure layer between your agents and this server.