Skip to main content
Glama

compile_ai_sandbox_policy

Generate zero-lateral-movement firewall and ACL policies for AI coding sandboxes across nftables, pf, Tailscale ACL, or iptables, restricting access to approved API targets.

Instructions

Compile zero-lateral-movement firewall / ACL policy for autonomous AI coding agents (Linux nftables, macOS pf, Tailscale ACL JSON, or Linux iptables).

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "compile_ai_sandbox_policy"
  }
}

Input Schema

NameRequiredDescriptionDefault
backendNonftables
sandbox_ipsNo
allowed_api_targetsNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

C2.9/5.0
Behavior2/5

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

With no annotations supplied, the description carries the full disclosure burden and does almost none of it. 'Compile' implies producing a policy artifact, but it never says whether the result is returned inline, written to disk, or applied to a live firewall — a material difference for a security-control tool. Nothing is said about privilege requirements, side effects, or whether an empty/omitted input set yields an open or closed policy.

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?

A single front-loaded sentence that names the capability first and the backend variants parenthetically, with no filler. It is efficient, though the density means several distinct facts (purpose, audience, four formats) are compressed into one clause chain at the cost of any operational detail.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness2/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a three-parameter generator with zero annotation coverage and no output schema, the description is too thin. It does not explain the relationship between `sandbox_ips` and `allowed_api_targets` (i.e. what traffic the compiled policy permits vs. denies), nor does it describe the shape of the emitted policy — things the agent cannot recover from structured fields because none of them document behavior.

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 coverage is 0%, so the description must compensate for three undocumented parameters. It partially does for `backend` by mapping enum values to platforms ('Linux nftables, macOS pf, Tailscale ACL JSON, Linux iptables'), confirming they are output dialects rather than input sources. However, `sandbox_ips` and `allowed_api_targets` — the two parameters that actually determine the policy's semantics — are never mentioned, leaving the agent to guess their format (CIDRs vs bare IPs, hostnames vs URLs).

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?

States a specific verb ('Compile') and resource ('zero-lateral-movement firewall / ACL policy') plus the target consumer ('autonomous AI coding agents'), which is enough for an agent to recognize the artifact being produced. It also enumerates the four supported backends, tying the output format to each platform. It stops short of naming or contrasting the sibling tools, but the purpose itself is unambiguous.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

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

There is no when-to-use guidance, no prerequisites, and no statement about when another tool (e.g. the audit or render siblings) is the better choice. The agent must infer from the tool name alone that this is the generation step performed after topology auditing. Nothing signals whether inputs are required or optional, so the agent cannot tell when this tool is even callable with a safe default.

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