OTP

Capture an operational learning

capture_learning

Capture an operational learning when an agent fails or you discover a better approach. This is the core OTP feedback loop: agent fails → you correct → correction becomes coordination intelligence → all agents improve. The learning is saved as a claim in your draft OOS and will be available via get_my_rules.

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": {
      "what_failed": "<what_failed>",
      "what_to_do": "<what_to_do>"
    },
    "name": "capture_learning"
  }
}

Input Schema

Name	Required	Description
`why`	No	Why this matters. E.g. 'Without date validation, the analysis includes pre-launch zeros that skew averages down'
`agent`	No	Which agent failed. E.g. 'Dash', 'Pepper', 'Dirk'
`section`	No	Which OOS section this belongs in (auto-detected if not specified)
`source_url`	No	URL to a better approach, repo, or reference. E.g. 'https://github.com/user/better-meta-analysis'
`what_to_do`	Yes	The rule/fix. E.g. 'Always validate date ranges against the campaign's actual active period before running spend analysis'
`what_failed`	Yes	What went wrong. E.g. 'Dash gave bad Meta Ads analysis -- used wrong date range and missed spend anomalies'

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 30, 2026

TDQS

A3.9/5.0

Behavior4/5

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

Annotations already flag this as a non-read-only, non-destructive, closed-world write, so the lower bar applies. The description adds genuine blast-radius context the annotations cannot: the result is stored as a claim in the caller's draft OOS and becomes retrievable by other agents via get_my_rules. It stays silent on dedup/idempotency and on whether the learning is immediately live or requires publishing.

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 trigger, then the loop rationale, then the storage destination. The arrow-chain sentence is slightly promotional but does convey causality efficiently; no sentence is pure filler.

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

Completeness4/5

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

With no output schema, the description correctly closes the loop by naming where the learning lands (draft OOS claim) and how it is read back (get_my_rules). A mutation tool with full schema coverage and annotations needs little more; the only real gap is the draft-vs-published lifecycle.

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?

Schema description coverage is 100% and every one of the six parameters carries its own example, so the schema does the heavy lifting. The description adds no format, length, or content guidance beyond what the schema already provides; baseline 3 applies.

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 and resource (capture an operational learning) plus the trigger conditions: agent failure or a discovered better approach. That distinguishes it semantically from the nearby add_issue/add_todo write tools, though it never names them directly.

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

Usage Guidelines4/5

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

Gives clear when-to-use conditions ('when an agent fails or you discover a better approach') and points to get_my_rules as the retrieval path, which is real routing value. No explicit when-not or exclusion criteria are offered.

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

All OTP 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?