Skip to main content
Glama

record_replay

Replay a recorded JSONL trajectory to verify each step as verified, unverified, or failed; use plan-only mode to preview actions or live mode to send real input and judge against expectations.

Instructions

Replay a recorded trajectory (JSONL) and report, per step, whether it was verified / unverified / failed. live=false (the default) sends no input: the actions come back as 'planned'. live=true delivers real keyboard and mouse input and judges each step against the expectation recorded in the trajectory. 'sent' only means delivered - it is not proof of effect, and a step with no recorded expectation stays 'unverified'. Failed steps keep their reason and error in actions[] and in failed_steps[], together with the step's 'activation' receipt (null when no activation was attempted).

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": {
      "trajectory_path": "<trajectory_path>"
    },
    "name": "record_replay"
  }
}

Input Schema

NameRequiredDescriptionDefault
liveNofalse = plan only and send nothing; true = send real input and verify
retriesNoattempts per step when the expectation is judged
timeoutNoseconds to wait for the expectation to hold
activateNolive only: bring each step's window to the foreground before delivering. Real input lands on whatever is in front, so without it a background target fails. Steals focus. Ignored when live=false: a dry run never moves a window.
intervalNoseconds between probes
trajectory_pathYespath to a trajectory.jsonl (header + one step per line)

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.2.0

TDQS

A4.3/5.0
Behavior5/5

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

With no annotations, the description carries the full behavioral burden and does so: it discloses the default dry-run, the caveat that 'sent' is not proof of effect, that steps lacking a recorded expectation stay 'unverified', and that failures carry reason/error plus an 'activation' receipt (null if none was attempted). This is exactly the context an agent needs before choosing live vs plan.

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?

Dense but front-loaded: purpose first, then the mode distinction, then result semantics and the 'sent' caveat. Every sentence carries information an agent would otherwise have to guess, with no 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?

For a 6-parameter tool with no output schema and no annotations, the description covers modes, status vocabulary, and the shape of failure reporting (actions[], failed_steps[], activation receipt), which is sufficient to call it correctly. It does not discuss side-effect risk of live input beyond focus stealing (which lives in the schema) or the retry/timeout interaction in prose.

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 100%, so the baseline is 3; the description goes further by explaining what live=true actually does at verification time and by grounding the 'expectation' concept used by retries/timeout. It adds meaning about the live parameter's consequences rather than merely restating types.

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?

Opens with a specific verb and resource ('Replay a recorded trajectory (JSONL)') and states the output shape (per-step verified/unverified/failed), so the agent knows exactly what it gets back. It does not name a contrasting sibling such as act_and_verify, so the differentiation is implicit rather than explicit.

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?

Clearly explains the two operating modes: live=false (default) plans only and sends nothing, live=true delivers real input and verifies against recorded expectations. No explicit when-to-use-this-vs-alternatives guidance or prerequisites are given, which keeps it out of the top band.

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