Skip to main content
Glama

Guided Assessment

guided_assessment
Destructive

Plans and guides security assessments by auto-detecting the right workflow and tools, returning triage, recommendations, and next steps before any command runs. Optionally solves tasks autonomously.

Instructions

Plan, guide, or autonomously solve a security task over the MCP toolchain.

An orchestrator on top of the registry, advisors, install checks, audit logging, and execution policy. Bootstrap commands use the governed execute_tool() path, so target scope, external-network, shell-injection, and blocked-flag checks apply.

By DEFAULT it auto-detects the right workflow + tools for the problem (workflow/ target_type="auto") and acts as a companion: it returns classification, triage gates, recommended skills, reporting next steps, a plan, tool install status, next actions, and the full MCP toolchain surface WITHOUT auto-running commands in this initial call. The agent can then run tools step by step as the user approves. The heaviest mode (autonomous) starts the auto-solver contract: it bootstraps triage, then the client agent continues with the full MCP toolchain (registry/advisors/install checks/run_tool/run_pipeline and separately gated run_script). When registry tools and pipelines are not enough, autonomous mode may create, save, and run scoped helper scripts for the user, persisting reusable ones under manual_scripts/. Simple recon/HTTP commands such as curl remain run_tool calls.

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": {
      "target": "<target>"
    },
    "name": "guided_assessment"
  }
}

Input Schema

NameRequiredDescriptionDefault
modeNo"companion" (default) or "autonomous" (opt-in).companion
targetYesURL, hostname/IP, or local file path to assess.
findingNoOptional short finding summary to classify for triage/report routing. Raw finding text is used locally but not echoed in the result.
workflowNo"auto" (default — inferred), "bounty", "ctf", or "generic".auto
intensityNo"low" (default) or "medium". Medium may include low-volume nmap.low
max_stepsNoMaximum number of bootstrap steps autonomous mode auto-executes.
target_typeNo"auto" (default — inferred from the target) or an explicit type: bounty type (web_app/api/cloud/network/iot/mobile_app) or CTF category.auto
authorization_confirmedNoRequired before any network step executes.

Output Schema

NameRequiredDescriptionDefault

No arguments

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv1.2.0

TDQS

A4.1/5.0
Behavior5/5

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

Beyond the destructive/openWorld annotations, the description discloses that bootstrap commands go through a governed execute_tool path subject to target scope, external-network, shell-injection, and blocked-flag checks, that authorization_confirmed gates network steps, and that the default call runs nothing automatically. It further discloses that autonomous mode may create, save, and persist reusable helper scripts under manual_scripts/, which is critical behavior not implied by any annotation.

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?

Purpose is front-loaded, but the prose is heavy, repeats 'MCP toolchain' and 'registry/advisors/install checks' multiple times, and packs several clauses into run-on sentences with awkward line breaks. Every sentence carries some information, yet the same mode behavior is restated twice (companion default, autonomous contract).

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 complex orchestrator with 8 parameters and an existing output schema, the description covers the two operating modes, what the default call returns, the security gates, and the escalation path to saved scripts. Remaining gaps are minor since return values are documented by the output schema and parameters are fully described in the schema.

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, but the description adds semantic weight: the mode parameter is framed as a safety/weight tradeoff (companion default vs autonomous opt-in), workflow and target_type are described as auto-inferred by default, and max_steps is tied to the number of bootstrap steps autonomous mode auto-executes. This goes modestly beyond restating field docs.

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 first sentence states a specific verb set and resource: 'Plan, guide, or autonomously solve a security task over the MCP toolchain,' and the second positions it as an orchestrator over registry, advisors, install checks, audit logging, and execution policy. This distinguishes it from flat siblings like run_tool or check_installed. It is dense and lacks a single crisp one-liner, which keeps it short of 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 Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It clearly states the default path (companion mode with workflow/target_type='auto', returning plans and recommendations without auto-running commands) versus the opt-in heavy path (autonomous mode bootstrapping triage). It also gives an explicit routing exclusion: 'Simple recon/HTTP commands such as curl remain run_tool calls,' and reserves helper scripts for when registry tools and pipelines are not enough. No explicit 'do not use when' list beyond those examples.

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