Skip to main content
Glama
brunoboto96

SceneScout

by brunoboto96
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

scout_run_plan

Run up to 20 UI actions in one call to complete forms or wizards, resolving targets by semantic locators and stopping at the first failed step or policy violation.

Instructions

Execute up to 20 actions in ONE call — use for mechanical sequences (fill a form, walk a wizard) so each step doesn't cost a round-trip. Targets resolve at execution time by semantic locator: 'testid=…', 'text=…', 'label=…' or 'role=button[name=Save]' (never snapshot refs). The same steps, saved to .scenescout/flows/.json with expect-text / expect-url / expect-request steps added, are replayed by scenescout check on every pull request. An upload step attaches a file as scout_upload does (target required — the file input or the control that opens its chooser; value = a fixture kind or a project-relative path). The plan ABORTS at the first NEW oracle violation, policy refusal, or failed step, returning a transcript of how far it got; repeats of already-reported violations do not abort (they stay logged for the report). For a sweep of independent steps (tabs, filters, pages) pass onViolation "continue": a new error status or its console echo is listed on its step's line and the plan goes on; a failed step, a policy refusal or any other violation still stops it. A select step whose value names no option fails at once, listing the options.

Input Schema

NameRequiredDescriptionDefault
taskNoWhat you are DOING right now, in a few words: the action, not the acceptance criteria. "Filtering the documents register by status", "Filling the deviation form with invalid dates", "Signing in as QA_Team" — NOT "§2.4 filtering narrows the set and the filter is reflected in the URL", which is what you are CHECKING, not what you are doing. Naming the item you are on is fine ("§2.4: filtering the documents register"); keep the rest to what a colleague would see over your shoulder. It stays set until you pass a different one, so a batch costs a few words, not one per call. Required on the tools that act unless a journey or an earlier call already set one.
stepsYes
sessionNoTarget this session directly instead of the active one — pass it explicitly when dispatching to MULTIPLE sessions in one turn (e.g. two scout_click calls with different `session`), which then run CONCURRENTLY rather than queueing. Omit for single-session sequential use.
objectiveNoOld name for `task` (2.0). Prefer `task`.
onViolationNostop (default): end the plan at the first new oracle violation, right for a form flow whose steps depend on each other. continue: list a new http_error or console_error on its step's line and run the next step, for a sweep of independent stepsstop

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changedv3.31.0
    • removedInput schema / additionalProperties
      Removed value: -false
    • removedInput schema / properties / steps / items / additionalProperties
      Removed value: -false
  2. Changed1 schema field changedv3.17.0
    • addedInput schema / properties / onViolation
      Added value: +{
+  "default": "stop",
+  "description": "stop (default): end the plan at the first new oracle violation, right for a form flow whose steps depend on each other. continue: list a new http_error or console_error on its step's line and run the next step, for a sweep of independent steps",
+  "enum": [
+    "stop",
+    "continue"
+  ],
+  "type": "string"
+}
  3. Changed1 schema field changedv3.11.1
    • changedInput schema / properties / steps / items / properties / target / description
      Previous value: -"testid=…, text=…, label=… (or a path for navigate; 'top'/'bottom'/±px for scroll; for upload: the file input or the control that opens its chooser)"New value: +"testid=…, text=…, label=…, role=<role>[name=\"…\"] (or a path for navigate; 'top'/'bottom'/±px for scroll; for upload: the file input or the control that opens its chooser)"
  4. Changed2 schema fields changedv3.4.0
    • addedInput schema / properties / objective
      Added value: +{
+  "description": "Old name for `task` (2.0). Prefer `task`.",
+  "maxLength": 120,
+  "type": "string"
+}
    • addedInput schema / properties / task
      Added value: +{
+  "description": "What you are DOING right now, in a few words: the action, not the acceptance criteria. \"Filtering the documents register by status\", \"Filling the deviation form with invalid dates\", \"Signing in as QA_Team\" — NOT \"§2.4 filtering narrows the set and the filter is reflected in the URL\", which is what you are CHECKING, not what you are doing. Naming the item you are on is fine (\"§2.4: filtering the documents register\"); keep the rest to what a colleague would see over your shoulder. It stays set until you pass a different one, so a batch costs a few words, not one per call. Required on the tools that act unless a journey or an earlier call already set one.",
+  "maxLength": 120,
+  "type": "string"
+}
  5. Addedv1.1.0

TDQS

A4.6/5.0
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 states the plan aborts at the first NEW violation, policy refusal or failed step, that repeat violations do not abort, that 'continue' only tolerates new http_error/console_error while failed steps and policy refusals still stop, and that a bad select value fails immediately listing options. It also discloses that targets resolve at execution time and that plans can be replayed by `scenescout check` in CI.

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 one-call/20-action hook and the use case, then ordered by importance: locators, CI replay, upload, abort semantics, continue mode. It is dense and long for a description, with some detail (the `scenescout check` PR-replay digression) that is orientation rather than invocation guidance, but almost no sentence is filler.

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?

No output schema exists, and the description compensates by describing what comes back on failure ('returning a transcript of how far it got') and how errors are surfaced per step. For a 20-step batch tool with no annotations, the failure modes an agent must reason about (abort vs. continue, select-option mismatch, upload target requirement) are all covered.

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 80%, so the schema already documents most fields; the description still adds real meaning — execution-time semantic locator resolution and the 'never snapshot refs' prohibition, and that upload's value is a fixture kind or project-relative path with a required target. It adds nothing for `task`, `objective` or `session` beyond what the schema says.

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?

Opens with a specific verb, scope and batching rationale: 'Execute up to 20 actions in ONE call'. This unambiguously separates it from the single-action siblings (scout_click, scout_type, scout_select, scout_hover) and from scout_guided_plan / scout_recheck_plan.

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 a clear trigger ('mechanical sequences (fill a form, walk a wizard)') and an explicit alternative condition for the onViolation "continue" mode ('a sweep of independent steps (tabs, filters, pages)'). It does not explicitly say to prefer scout_click for one-off actions or reference the sibling plan tools by name, so routing vs. scout_guided_plan / scout_journey is left implicit.

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

Deploy Server

Other Tools