Opens a write session and returns a session_id, enabling agents to record code changes and relationships in the Kin repository.
Session plumbing for writes. Open one; get a session_id.
| Name | Required | Description | Default |
|---|---|---|---|
| cwd | Yes | Agent's cwd. | |
| pid | No | Agent process id. | |
| vendor | Yes | e.g. claude-code, codex. | |
| transport | No | mcp|cli|wrapper|ui | mcp |
| session_id | No | Optional UUID to register under. | |
| client_name | Yes | Client name. | |
| capabilities | No | Abilities |
Changes observed during successful MCP inspections.
v0.7.17Input schema / properties / capabilities / descriptionPrevious value: -"Agent capabilities"New value: +"Abilities"Input schema / properties / client_name / descriptionPrevious value: -"Human-readable client name"New value: +"Client name."Input schema / properties / cwd / descriptionPrevious value: -"Working directory of the agent"New value: +"Agent's cwd."Input schema / properties / pid / descriptionPrevious value: -"OS process ID of the agent (optional)"New value: +"Agent process id."Input schema / properties / session_idAdded value: +{
+ "description": "Optional UUID to register under.",
+ "type": "string"
+}Input schema / properties / transport / descriptionPrevious value: -"Connection type: mcp, cli, wrapper, or ui"New value: +"mcp|cli|wrapper|ui"Input schema / properties / vendor / descriptionPrevious value: -"Vendor identifier (claude-code, codex, gemini-cli, etc.)"New value: +"e.g. claude-code, codex."v0.7.16Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare the safety profile (not read-only, not destructive, not idempotent, not open-world), so the description's bar is lower. The description adds that it is for writes and returns a session_id, but it does not disclose session lifecycle details, reuse behavior, or what happens if capabilities are omitted.
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 description is two short sentences with no filler and front-loads its purpose. It is efficient, though its telegraphic style may be overly terse for a tool with seven parameters and a nested capabilities object.
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 session-start tool with seven parameters, a nested capabilities object, no output schema, and many session/transaction siblings, the description is too sparse. It does not cover session prerequisites, expiration, capability defaults, or relationship to kin_transaction_begin, leaving key operational context unaddressed.
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%, so all seven parameters, including the nested capabilities object, are already documented in the schema. The description adds no parameter-level meaning, which is acceptable given the high schema coverage baseline of 3.
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 action ('Open one') and resource ('session'), and notes the return value ('get a session_id'). It also gives scope ('for writes'), though it does not distinguish this tool from siblings like kin_transaction_begin or kin_session_end.
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 implies usage for write operations via 'Session plumbing for writes' and the imperative 'Open one; get a session_id.' However, it never explicitly states when to choose this tool over alternatives, such as kin_transaction_begin or kin_session_exec, leaving the agent to infer the sequencing.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.