build_project_kg

THE RISK

Medium Risk

Why build_project_kg needs a policy

Building a per-project canonical knowledge graph layer is a Write operation: it creates and stores structured data. It is reversible in principle (the KG can be rebuilt or deleted), so it doesn't reach Destructive. The blast radius is medium — a misconfigured KG could corrupt architecture recommendations for the project, but the scope is limited to a single project's consulting data.

From the tool's definition 'BUILD PROJECT KG — Build the per-project canonical layer' — creates/builds a knowledge graph structure for a project

Documented attack patterns abuse exactly the kind of access build_project_kg gives an agent:

POLICY

How to control build_project_kg

PolicyLayer is an MCP gateway — it sits between your AI agents and Iconsult MCP, and nothing reaches the server without passing your rules. This is the rule we recommend for build_project_kg:

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "build_project_kg": {
      "limits": [
        {
          "counter": "build_project_kg_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

build_project_kg stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

Create a free account and register Iconsult MCP — nothing to install.
Add this policy — paste it, or build it visually.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

LIMIT THIS TOOL →

Free to start. No card required.

EXPLORE

Related tools and policies

More Iconsult MCP tools

Execute emit_event EVENT (emit) — Emit a consultation event for reactive processing. Write start_project START PROJECT — Create or refresh a per-project cache row. Write assert_fact BLACKBOARD (assert) — Assert a typed, versioned fact on the blackboard. Write log_pattern_assessment LOG ASSESSMENT — Record a pattern assessment for a consultation session. Write rate_consultation QUALITY (rate) — Rate a consultation Write generate_implementation_plan IMPLEMENTATION PLAN — Generate a phased, classified implementation plan Write plan_consultation PLAN — Generate an adaptive consultation plan after match_concepts. Write update_plan_step UPDATE STEP — Update the status of a step in an implementation plan.

All 29 Iconsult MCP tools →

Write tools on other servers

M-Team MCP Server download_torrent YouTube MCP Server generate_video_title Android Forensics ADB MCP Server combine_reports 0xarchive web3_signup

Go deeper

The MCP Attack Database → Documented attack patterns against MCP deployments — and the policies that stop them.
Argument validation → Checking tool call arguments against schema and policy before they run.
Scoped token → Per-person credentials that grant a defined subset of servers and tools.
Rate Limiting MCP Tool Calls: A Practical Guide →
MCP Security: Why Prompt Guardrails Aren't Enough →

FAQ

Questions about build_project_kg

What does the build_project_kg tool do? +

BUILD PROJECT KG — Build the per-project canonical layer for. It is categorised as a Write tool in the Iconsult MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on build_project_kg? +

Register the Iconsult MCP server in PolicyLayer and add a rule for build_project_kg: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Iconsult MCP. Nothing to install.

What risk level is build_project_kg? +

build_project_kg is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit build_project_kg? +

Yes. Add a rate_limit block to the build_project_kg rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block build_project_kg completely? +

Set action: deny in the PolicyLayer policy for build_project_kg. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides build_project_kg? +

build_project_kg is provided by the Iconsult MCP server (marcus-waldman/iconsult_mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Iconsult MCP tool call.

Start from Iconsult MCP, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

CHECK YOUR STACK →