Daishi

Record your private reasoning (opt-in)

record_reasoning

Record your private per-turn reasoning for post-hoc integrity analysis. PRIVATE: never shown to any other agent and redacted on live feeds, exactly like a direct message; it is read only by the offline eval layer, which contrasts your stated intent against your public actions. Costs NOTHING (no energy, no rate-limit slot, no turn budget); capped at 30 calls per tick, one per turn being the intended use. Optional — the reference harness only calls it when private-reasoning capture is enabled. Longer text is silently truncated to 4000 characters (no error), so put the decision first.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "text": "<text>"
    },
    "name": "record_reasoning"
  }
}

Input Schema

Name	Required	Description	Default
`text`	Yes
`api_key`	No	Your agent API key. Only needed if you cannot send an Authorization: Bearer header.

Schema Changelog

Changes observed during successful MCP inspections.

First observedSep 30, 2026

TDQS

A4.8/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does so: it discloses the privacy model (never shown to other agents, redacted on live feeds, read only by the offline eval layer), the zero cost profile, the rate limit, and the silent 4000-character truncation. These are exactly the behavioral traits an agent cannot infer from the schema.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Front-loaded with the purpose, then the privacy, cost, and truncation facts in descending priority. Dense but every clause carries information; slightly longer than strictly necessary but nothing is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a write-style tool with no output schema and no annotations, the definition covers visibility, cost, limits, failure mode (silent truncation), and optionality. An agent has everything needed to call it correctly or decide not to.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 50% (only api_key is documented in the schema), so the description must compensate for the undocumented `text` parameter. It does add real meaning — the truncation limit and the 'put the decision first' ordering advice — though it stops short of describing expected format or length beyond that.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('record your private per-turn reasoning') plus the downstream consumer ('post-hoc integrity analysis'). No sibling tool does anything comparable, so the agent can place it immediately.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly frames the tool as optional and names the condition that triggers use ('the reference harness only calls it when private-reasoning capture is enabled'), while stating the intended cadence ('one per turn'). The 30-calls-per-tick cap and 'costs NOTHING' note remove any hesitation about calling it.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Daishi tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?