Skip to main content
Glama

Export Recon Kit

export_recon_kit
Idempotent

Builds a recon kit for binaries, archives, firmware, or AI models, producing a blueprint, interactive explorer, typed artifacts, and ready-to-run engineering handoff prompts.

Instructions

Materialize an ArgonZeroG Universal 4-Mount Recon Kit (RECON_BLUEPRINT.md, generative_ui RECON_EXPLORER.html interactive workbench, and typed recon-kit/ artifacts) for any native binary (ELF/PE/Mach-O, C++ RTTI/VTables, stripped Rust), packaged UI/game archive (Electron ASAR, Unity IL2CPP, Godot .pck, APK/DEX, Web/SourceMap, PE resources), hardware/firmware/protocol artifact (SVD/DBC/UF2/PCAP/Sigrok), or AI model/data artifact (GGUF/SafeTensors/ONNX/TFLite/SQLite) with ready-to-run Harness 1..4 engineering handoff prompts.

Example tool call

This call uses example arguments supplied in the tool’s input schema.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "path": "/artifacts/target_binary.exe",
      "max_classes": 500,
      "max_entries": 500,
      "max_subsystems": 50,
      "mount_override": "auto",
      "mock_ui_concepts": false,
      "include_explorer_html": false
    },
    "name": "export_recon_kit"
  }
}

Input Schema

NameRequiredDescriptionDefault
pathNoFilesystem path to the target binary, archive, firmware image, capture, or AI model artifact (alias for target_path).
output_dirNoOptional output directory where recon-kit/, RECON_BLUEPRINT.md, and RECON_EXPLORER.html will be materialized (defaults to <target_dir>/<target_stem>-recon-kit).
max_classesNoMaximum number of recovered C++/Rust/IL2CPP/.NET classes or types to materialize (default 500).
max_entriesNoMaximum number of archive entries, functions, tensors, or hardware registers to sample per mount (default 500).
target_pathNoFilesystem path to the target binary, archive, firmware image, capture, or AI model artifact.
max_subsystemsNoMaximum number of clustered subsystems/components to emit in the blueprint and explorer (default 50).
mount_overrideNoMount selection mode: 'auto' (default, automatically detects all applicable mounts via magic bytes and internal structures) or an explicit mount override.
mock_ui_conceptsNoOpt-in flag (default false): when true and the target contains UI/UX assets, renders an Antigravity generative_ui visual concept mock of the extracted screens/components without any clickable harness buttons.
include_explorer_htmlNoOpt-in alias for mock_ui_concepts (default false): no HTML files are generated unless explicitly requested by the user.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.3/5.0
Behavior3/5

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

Annotations already declare readOnlyHint=false and idempotentHint=true, and the description is consistent with that by saying it materializes files on disk with a default output directory convention. It adds some useful context (which three artifact families get written, that Harness 1..4 prompts come along), but says nothing about runtime cost, partial-failure behavior, or how much disk a large mount consumes – the missing pieces for an expensive generation tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The verb-first opening is well front-loaded, but the whole definition is a single ~85-word run-on sentence built from parenthetical enumerations of formats. A reader has to parse an artifact taxonomy to reach the three things actually produced; splitting the output artifacts into their own sentence would improve it without adding length.

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?

For a 9-parameter, no-output-schema, multi-format generation tool with only safety-related annotations, the description covers inputs and the three output artifacts but omits return-value shape, cost/runtime expectations, and how the four mount families interact with the mount_override parameter. It is adequate to select the tool but thin for predicting what the call will actually produce.

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%, so every parameter is already documented in structured form, including defaults, ranges, and the mount_override enum-free string semantics. The description adds no parameter-level guidance of its own (nothing on when to raise max_classes or when to force a mount override), so the 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?

The verb 'Materialize' is paired with a concrete named artifact set (RECON_BLUEPRINT.md, RECON_EXPLORER.html, recon-kit/ artifacts) and an explicit input domain, so an agent can tell this is a kit-generation tool rather than a comparison or single-artifact export. It does not explicitly contrast itself with the likelier confusable siblings (export_evidence_bundle, export_web_scripts), which keeps it off a 5.

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 long input enumeration ('any native binary... packaged UI/game archive... hardware/firmware... AI model') implicitly defines when the tool applies, and the mention of Harness 1..4 handoff prompts hints at downstream use. But there is no explicit when-not condition, no named alternative for narrower tasks like binary_overview or export_evidence_bundle, and no prerequisites (e.g., does the target need to be opened in a session first?).

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