Create a new Open Knowledge Format bundle by generating a root index.md with a document list and log.md, refusing if index.md exists, then add typed Markdown concepts and lint them.
Start a bundle: a root index.md declaring okf_version "0.2" with a generated document list, and a
§9 log.md. Refuses if index.md exists. Then add concepts (any .md with type: frontmatter) and lint.
| Name | Required | Description | Default |
|---|---|---|---|
| apply | No | ||
| bundle | No |
Changes observed during successful MCP inspections.
v0.1.0Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=false, destructiveHint=false and idempotentHint=true, and the description usefully adds the refusal precondition and the exact files provisioned, which explains why it is non-destructive. However it never discloses that `apply` defaults to false, i.e. that a default call is effectively a dry run and no files are written unless apply is set — a major behavioral trait for a scaffolding tool.
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?
Dense and front-loaded: the core action and artifacts come first, followed by the precondition and the next-step workflow. Every clause carries information; the only rough edge is the abrupt 'and lint' fragment.
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?
No output schema exists, so the description must convey the result — and it does, by naming the exact files created. Combined with the refusal condition, the descriptions of a two-parameter, non-destructive write tool are essentially complete; the missing dry-run/apply semantics are already penalized under parameter semantics rather than being a separate completeness gap.
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?
Both parameters have 0% schema description coverage, so the description carries the full burden, yet neither `apply` (dry-run vs. commit) nor `bundle` (target path/identifier) is mentioned or explained. The only hint at semantics is contextually implied by 'Refuses if index.md exists'; an agent could easily call this expecting files to be created when apply defaults to false.
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-resource pair ('Start a bundle') and enumerates the concrete artifacts produced (root index.md with okf_version "0.2", generated document list, §9 log.md). That level of artifact detail reliably distinguishes it from siblings like create, index, and upgrade, even without naming them explicitly.
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 a clear precondition ('Refuses if index.md exists') and a follow-on workflow ('Then add concepts... and lint') that routes the agent to the concept-creation and linting tools. It stops short of naming the sibling tools by name, so the routing is implied rather than explicit.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.