Discover Drawcall workflows for 3D apps and asset creation; read APIs and approve CLI logins.
Scored across 3 tools
approve_login is clearly separate (authentication), and search_skills vs read_skills form a conventional find-then-read pair whose descriptions explicitly state search returns paths and read_skills returns full guides. The boundary is understandable, though an agent could initially reach for read_skills without a path from search.
All three tools follow a clean snake_case verb_noun pattern: approve_login, read_skills, search_skills. The verbs (approve/read/search) accurately map to the distinct actions.
Three tools is lean but each earns its place: one auth action plus a search/read pair for the documentation surface. It is on the thin side, but the server's scope is narrow (skill discovery plus CLI login) rather than a full CRUD domain.
Discovery and reading of workflows are covered, along with the login approval flow, but there is no tool to list packages or directly inspect Market inventory/assets, and no logout/revoke counterpart to approve_login. Agents must chain search_skills into read_skills and rely on external CLI commands for anything beyond documentation lookup.
Approve a pending Drawcall CLI login as your connected account. This grants the CLI a full account session. Only approve the user_code printed by the user's current drawcall auth login command. After approval, run drawcall auth complete in that CLI.
| Name | Required | Description | Default |
|---|---|---|---|
| user_code | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations declare a non-readOnly, non-idempotent, non-destructive mutation, and the description adds meaningful context beyond them: the operation 'grants the CLI a full account session,' which is a security-relevant consequence worth surfacing. It does not detail failure behavior for an invalid or expired code, leaving a minor gap.
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 the action, each contributing distinct information (effect, precondition, follow-up step). No repetition of the title or wasted phrasing.
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 single-parameter mutation tool with no output schema, the description covers purpose, security consequence, parameter origin, and the required follow-up command. Missing only edge-case behavior (what happens with an already-used or invalid code), which is minor given annotation coverage of safety traits.
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 0%, so the description carries the burden for user_code. It helpfully identifies the value's origin ('printed by the user's current drawcall auth login command') but adds no format, length, or validity details, and the schema only declares it as a bounded string.
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 ('Approve a pending Drawcall CLI login') and immediately clarifies the scope via 'as your connected account.' An agent can tell exactly what action occurs, and the sibling tools (read_skills, search_skills) are unrelated, so no differentiation burden exists.
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?
Provides clear procedural context: approve only the user_code from the user's current 'drawcall auth login' command, then run 'drawcall auth complete'. This is effectively a when-to-use plus an implicit when-not ('Only approve...'). No named alternative tool exists to route against, so a 4 is appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Read a guide for its workflow and prerequisites, or a declaration for API details. When the workflow needs user action, give the documented command or URL as the next step. A package path lists its skills and declaration sources; a SKILL.md gives its full guide. Paths come from search results and read_skills listings; listing names are relative to the path that was read.
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes | A path from search or a listing, e.g. /@drawcall/physics/ |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnly, idempotent, non-destructive, open-world behavior, so the safety profile is covered. Beyond that, the description discloses return-shape behavior that no schema documents: a package path yields a skill listing plus declaration sources, a SKILL.md yields the full guide, and listing names are relative to the read path. Error handling and failure modes are not covered.
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?
The verb and resource lead the description, and the four sentences are dense with no filler. The second sentence about user action is slightly tangential to a read operation but still carries actionable instruction, so the text earns most of its length.
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?
With no output schema, the description carries the burden of explaining return values, and it does distinguish the two result shapes (listing versus full guide). An agent has enough to call it correctly; only error or empty-result behavior is left unstated.
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 path parameter already carries an example, so the baseline is 3. The description earns above baseline by adding provenance semantics (paths originate from search or listings) and the non-obvious rule that listing names are relative to the path that was read, which the schema does not state.
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 (read a skill guide or declaration) and explains the two content types it returns: a workflow guide with prerequisites versus an API declaration. It doesn't explicitly contrast itself with search_skills by name, though the path-provenance sentence implies the relationship.
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 gives clear usage context: paths come from search results and read_skills listings, which routes the agent from search_skills to this tool, and it states what the agent should do next (surface the documented command or URL) when user action is required. There is no explicit 'do not use this for X' exclusion, which keeps it below a 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search Drawcall workflows and APIs for apps, games and assets. Discover asset sourcing as well as runtime libraries. Source visual content from @drawcall/market first. When its candidates do not fit, consult @drawcall/generate for generation options and prerequisites before settling on a fallback. Reserve hand-authored geometry for gameplay structure or explicitly procedural content. @drawcall/rig adds skeletons and skin weights when needed. This searches documentation, not Market inventory: read the guides to find and inspect actual assets. Search covers all supported packages automatically, including in an empty project. Returns up to five paths with excerpts; read_skills provides the full workflows and APIs.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | What you want to implement or learn how to do |
| Name | Required | Description |
|---|---|---|
| results | Yes | The five best, best first, at most one per document. Empty until every selected package is ready. |
| packages | Yes | Every selected package, in selection order |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover the safety profile (readOnly, idempotent, openWorld), yet the description still adds real context: it searches documentation rather than inventory, covers all supported packages even in an empty project, and caps results at five paths with excerpts. Return value detail is partly carried by the output schema, so 4 rather than 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?
The core purpose is front-loaded, but the middle is padded with package-routing advice (the @drawcall/rig sentence in particular feels tangential to how to invoke this search). Several sentences are dense and could be trimmed without losing selection-relevant 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?
With a rich output schema and strong annotations, the main missing piece is understanding scope, which the description covers (all packages, empty projects, docs vs inventory, result cap). An agent has enough to call it correctly; only the query-phrasing nuance is left implicit.
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?
There is a single parameter with 100% schema description coverage, which already defines the query as 'What you want to implement or learn how to do.' The description adds no syntax, format, or phrasing guidance beyond that, so 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 opens with a specific verb and resource ('Search Drawcall workflows and APIs') and clarifies the domain (apps, games, assets; documentation not Market inventory). It also distinguishes itself from the sibling read_skills by explaining that search returns paths and excerpts while read_skills returns full workflows.
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 gives explicit routing guidance among the @drawcall packages (market first, then generate, reserve hand-authored geometry, rig when needed) and states the escalation path to read_skills for full content. What it does not do is spell out a when-not-to-use-this-tool exclusion, so it stops short of a 5.
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.
search_skills1 field changedOutput schema / properties / results / items / properties / uses / descriptionPrevious value: -"For a returned section: paths of declaration sources it uses"New value: +"Declaration source paths relevant to the matched guide passage"search_skills1 field changedOutput schema / properties / results / items / properties / excerpt / descriptionPrevious value: -"The first lines: of a section its text, of a type its declaration"New value: +"Relevant guide text or a declaration from the public API and its supporting types"read_skills1 field changedInput schema / properties / path / descriptionPrevious value: -"e.g. /@drawcall/physics@0.1.0/types/RigidBody"New value: +"A path from search or a listing, e.g. /@drawcall/physics/"search_skills4 fields changedOutput schema / properties / packages / items / properties / status / descriptionPrevious value: -"ready: searched. indexing: not searched yet, search again shortly. failed: could not be indexed (see error). not_allowed: this service does not index the package."New value: +"ready: indexed and available. indexing: not indexed yet, search again shortly. failed: could not be indexed (see error). not_allowed: this service does not index the package."Output schema / properties / results / descriptionPrevious value: -"The five best, best first, at most one per document"New value: +"The five best, best first, at most one per document. Empty until every selected package is ready."Output schema / properties / results / items / properties / path / descriptionPrevious value: -"Where the full text is: a path of the tree, e.g. /@drawcall/physics@0.1.0/types/RigidBody"New value: +"Where the full text is: a path of the tree returned by this service"Output schema / properties / results / items / properties / uses / descriptionPrevious value: -"For a section: paths of the types it uses"New value: +"For a returned section: paths of declaration sources it uses"search_skills1 field changedInput schema / additionalPropertiesAdded value: +falsesearch_skills4 fields changedInput schema / properties / packagesRemoved value: -{
- "items": {
- "description": "An npm package as name@version (the installed version), or a bare name for its latest",
- "type": "string"
- },
- "maxItems": 200,
- "minItems": 1,
- "type": "array"
-}Input schema / properties / query / descriptionPrevious value: -"One concrete question, goal or symptom"New value: +"What you want to implement or learn how to do"Input schema / requiredPrevious value: -[
- "query",
- "packages"
-]New value: +[
+ "query"
+]Output schema / properties / packages / descriptionPrevious value: -"Every requested package, in request order"New value: +"Every selected package, in selection order"approve_loginread_skillssearch_skillsDiscover atmospheric 3D worlds, read controls, and build and publish authorized environments.
Free, read-only access to Alicia's published English and Chinese notes on Houdini, Three.js, OpenUSD and AI-assisted 3D workflows. Search articles, read full Markdown, browse topics and recent notes. Results include canonical source URLs and author attribution. No account or API key; Streamable HTTP with four read-only tools. Digital Garden is the interactive 3D flower artwork at digitalgarden.si. Setup: https://digitalgarden.si/agents/?utm_source=glama&utm_medium=referral&utm_campaign=listing
Build and inspect browser-based 3D games: projects, scenes, Lua scripts, assets and publishing.
Blender-as-a-service for agents: search 3D assets, run Blender Python, or brief the studio agent.
Add one secure layer between your agents and this server.