Skip to main content
Glama

oc_evidence_bundle

Capture a snapshot of the current page state including DOM, screenshot, network, and console, then bundle it into a directory for later analysis.

Instructions

Capture a snapshot of the current page state (DOM, screenshot, network slice, console, perceptual hash) and write it to a bundle directory. Returns { bundle_id, path, size_bytes, parts }. Default include = ['dom', 'screenshot']; pass include to capture more parts. network_window_ms (default 5000) limits the network slice to recent entries. Core-tier; does not depend on the pilot runtime.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
includeNoParts to capture. Default ['dom', 'screenshot']. Allowed: dom, screenshot, network, console, contract_facts, phash, gate, schema_diff. gate requires evidence.snapshot.gate; schema_diff requires target_schema and evidence.snapshot.observed. Missing inputs omit that part.
evidenceNoCaller-supplied snapshot. Provide the subset of fields the requested parts need: `dom`, `screenshot_png_base64`, `network`, `console`, `contract_facts`, `now_ms`, `gate`, `observed`. Missing fields cause the corresponding part to be omitted gracefully.
output_modeNo"inline" (default): return the full payload in-band — byte-identical to v1.11.0. "handle": write payload to the handle store and return a small descriptor; redeem with oc_output_fetch. "auto": inline if payload ≤ output_inline_limit_bytes, otherwise handle.
target_schemaNoDeclared target schema (see src/core/contracts/schema-diff.ts: { version: 1, fields: [ { name, type, required? } ] }). When supplied together with `evidence.snapshot.observed` and the 'schema_diff' part is included, the bundle writes `schema_diff.json` containing the structured field-match diff.
network_window_msNoRolling window (ms) used to slice the supplied `evidence.snapshot.network` array. Default 5000.
output_inline_limit_bytesNoOnly honored when output_mode="auto". If the serialized payload exceeds this byte count the response spills to a handle. Default: 32768.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changedv1.13.0
    • changedInput schema / properties / evidence / description
      Previous value: -"Caller-supplied snapshot. Provide the subset of fields the requested parts need: `dom`, `screenshot_png_base64`, `network`, `console`, `now_ms`, `gate`, `observed`. Missing fields cause the corresponding part to be omitted gracefully."New value: +"Caller-supplied snapshot. Provide the subset of fields the requested parts need: `dom`, `screenshot_png_base64`, `network`, `console`, `contract_facts`, `now_ms`, `gate`, `observed`. Missing fields cause the corresponding part to be omitted gracefully."
    • changedInput schema / properties / evidence / properties / snapshot / description
      Previous value: -"Snapshot fields. `dom` (string|object), `screenshot_png_base64` (base64 PNG), `network` (NetworkEntry[]), `console` (ConsoleEntry[]), `now_ms` (epoch ms used for the network window cutoff), `gate` (oc_gate_inspect-compatible fact), `observed` (structured extraction data for schema_diff)."New value: +"Snapshot fields. `dom` (string|object), `screenshot_png_base64` (base64 PNG), `network` (NetworkEntry[]), `console` (ConsoleEntry[]), `contract_facts` (bounded collector facts), `now_ms` (epoch ms used for the network window cutoff), `gate` (oc_gate_inspect-compatible fact), `observed` (structured extraction data for schema_diff)."
    • changedInput schema / properties / include / description
      Previous value: -"Which parts to capture. Default ['dom', 'screenshot']. Allowed items: 'dom' | 'screenshot' | 'network' | 'console' | 'phash' | 'gate' | 'schema_diff'. `gate` requires `evidence.snapshot.gate`; `schema_diff` requires `target_schema` and `evidence.snapshot.observed`; otherwise the part is omitted."New value: +"Parts to capture. Default ['dom', 'screenshot']. Allowed: dom, screenshot, network, console, contract_facts, phash, gate, schema_diff. gate requires evidence.snapshot.gate; schema_diff requires target_schema and evidence.snapshot.observed. Missing inputs omit that part."
    • changedInput schema / properties / include / items / enum
      Previous value: -[
      -  "dom",
      -  "screenshot",
      -  "network",
      -  "console",
      -  "phash",
      -  "gate",
      -  "schema_diff"
      -]New value: +[
      +  "dom",
      +  "screenshot",
      +  "network",
      +  "console",
      +  "contract_facts",
      +  "phash",
      +  "gate",
      +  "schema_diff"
      +]
  2. First observedv1.12.8

TDQS

A4.5/5.0
Behavior5/5

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

The description discloses key behaviors: it writes to a bundle directory, supports inline/handle/auto output modes, defaults the included parts, omits missing parts gracefully, and spills to a handle when auto limit is exceeded. It also references byte-identical v1.11.0 inline behavior, giving strong transparency.

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?

The description is detailed and dense, but the length is largely justified by the number of parameters and the nested evidence object. It could be slightly tightened, but it remains well-structured and front-loaded with the core purpose.

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?

Given the lack of an output schema, the description provides enough context: the return object shape, the handle redemption path via oc_output_fetch, the schema-diff format reference, and the behavior for missing evidence fields. An agent has sufficient information to invoke the tool and interpret its results.

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

Parameters5/5

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

All six parameters are explained with meaningful semantics beyond the schema. The description clarifies how include interacts with evidence fields, how network_window_ms slices the network array, how target_schema and observed produce schema_diff.json, and how output_inline_limit_bytes only applies in auto mode.

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?

The description clearly states the tool captures a snapshot of current page state (DOM, screenshot, network slice, console, perceptual hash) and writes it to a bundle directory. It also specifies the return shape, making the purpose concrete and distinguishable from related evidence tools.

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 description explains the mechanics and defaults well but does not explicitly contrast this tool with sibling tools such as oc_evidence_get, page_screenshot, console_capture, or network_capture. The intended use is implied by the bundling behavior but not explicitly stated as when to use it versus alternatives.

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