Aeroz

record_ai_decision

record_ai_decision

Destructive

Append a decision (or an incident) to an AI system's signed, append-only record, by the rules of POST /v1/ai/systems/{id}/records. Send the input and output text or their SHA-256: only the hash is kept, never the text. Nothing recorded can be edited or removed, so record what actually happened, not a guess.

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": {
      "system_id": "<system_id>"
    },
    "name": "record_ai_decision"
  }
}

Input Schema

Name	Required	Description
`kind`	No
`note`	No	Required for an incident.
`input`	No	The input, as text or JSON. Hashed, then dropped.
`output`	No	The output, as text or JSON. Hashed, then dropped.
`ref_id`	No	For an incident: the decision (air_…) it is about.
`subject`	No	What it was about: a unit, GTIN + serial, case id.
`system_id`	Yes	aisys_…, from list_ai_systems.
`confidence`	No	0 to 1.
`input_sha256`	No
`output_label`	No	A short readable label, e.g. suspect, clear.
`model_version`	No
`output_sha256`	No

Schema Changelog

Changes observed during successful MCP inspections.

AddedNov 9, 2026

TDQS

A3.9/5.0

Behavior5/5

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

Beyond the annotations, the description discloses the traits that matter most: the record is signed and append-only, nothing can be edited or removed, and input/output text is hashed and then dropped so only the SHA-256 is retained. That explains what the destructive/irreversible hint actually means in practice, which the annotations alone cannot convey.

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

Conciseness5/5

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

Three tight sentences that front-load the action, then the retention rule, then the caution about accuracy. No filler and nothing repeated.

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

Completeness3/5

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

For a 12-parameter, irreversible mutation with no output schema, the description covers intent and data-retention behavior well but defers the field-level rules to 'the rules of POST /v1/ai/systems/{id}/records' and never explains the decision/incident modes that drive several parameters.

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

Parameters3/5

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

It clarifies that 'input' and 'output' may be sent either as text/JSON or as a SHA-256 and that the text is discarded, which adds meaning for those two fields. With 67% schema coverage and 12 parameters, the remaining gaps (kind's decision-vs-incident distinction, note being incident-only, ref_id, confidence, subject) are left entirely to the schema.

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

Purpose4/5

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

States a specific verb ('Append') and a specific resource ('a decision or an incident to an AI system's signed, append-only record'), so the operation is unambiguous. It does not, however, contrast itself with the sibling review_ai_decision or register_ai_system, leaving the agent to infer which one writes versus reviews.

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

Usage Guidelines3/5

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

The line 'record what actually happened, not a guess' implies the tool is for factual post-hoc logging, which is useful context. But there is no explicit when-to-use vs when-not-to-use statement and no named alternative (e.g. review_ai_decision for reviewing an existing entry), so routing relies on inference.

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

All Aeroz 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?