Skip to main content
Glama

Simulate what the bandit would try

simulate_session

Replay the UCB1 policy against a tweak pool to see the order a fresh machine would try tweaks; optionally compare a cold start with pool-seeded behavior.

Instructions

Replays the UCB1 policy from the product against this pool and returns the order in which a fresh machine would try tweaks. ASSUMPTION: each measurement is assumed to return exactly the pool average for that tweak, because real per-run rewards are not stored. The order is therefore realistic; the rewards are not samples.

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": {
      "hardware_class": "<hardware_class>"
    },
    "name": "simulate_session"
  }
}

Input Schema

NameRequiredDescriptionDefault
roundsNo
use_poolNoSeed the bandit from the shared pool (true) or start cold, as the very first machine of its class would (false). Comparing the two shows what the pool is worth: cold, UCB1 must try every tweak once before it can prefer anything.
hardware_classYes

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A3.5/5.0
Behavior4/5

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

With no annotations, the description carries the burden and does well: it flags the core caveat that rewards are constant per-tweak averages rather than real samples, and warns that the ordering is realistic but the reward values are not. That is exactly the kind of limitation an agent needs to avoid misreporting results. It stops short of stating that the tool is a side-effect-free read of existing data, which is left to inference from 'simulate'.

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, front-loaded with the action, then the assumption in caps so it cannot be skimmed past, then the consequence. Nothing is padded and the ordering of information is deliberate.

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?

The return value is described (an ordering of tweaks), which matters since there is no output schema, and the reward caveat is covered. But for a tool whose required parameter is 'hardware_class', leaving that parameter's meaning entirely unexplained is a real gap for a 3-parameter tool.

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

Parameters2/5

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

Schema coverage is only 33%: use_pool is well documented in the schema, but 'hardware_class' (required) and 'rounds' (with its 1-100 bounds and default) are undocumented in both the schema and the description. The description never explains what a hardware class is here or what the round count controls, so it does not compensate for the coverage gap.

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?

Names a specific verb and resource: replays the UCB1 policy against the pool and returns the order a fresh machine would try tweaks. That is far more specific than the title alone. It does not, however, distinguish itself from the sibling 'recommend', which an agent could plausibly confuse with this simulation.

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?

Usage context is implied rather than stated: this is for inspecting how the policy would explore, and the use_pool description hints at a comparative cold-vs-pool experiment. There is no explicit when-to-use-this-instead-of-'recommend' or 'tweak_stats' guidance, and no stated prerequisites.

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