// SCAN REPORT

Project Graph

19 tools. 5 can modify or destroy data without limits.

5 write tools that can modify data. Rate limits recommended.

Last updated: 29 Jun 2026

5 can modify or destroy data

14 read-only

19 tools total

Community server · catalogue entry verified 29/06/2026

How to control Project Graph ↓

TOOL MAP

What Project Graph exposes to your agents

Read (14) Write / Execute (5) Destructive / Financial (0)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

High Risk

The most dangerous Project Graph tools

Write · Medium filters The tool modifies filter configuration settings (set, add_excludes, remove_excludes, reset). Write · Medium testing This tool manages a test checklist with actions to mark tests as pending, pass, or fail, and to reset the checklist. Write · High compact The tool performs reversible modifications to code (edit, compact, beautify operations). Write · Medium set_custom_rule This tool creates or modifies custom analysis rules within the project-graph system.

5 of Project Graph's 19 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Project Graph

PolicyLayer is an MCP gateway — it sits between your AI agents and Project Graph, and nothing reaches the server without passing your rules. These are the rules we recommend:

Rate limit write operations

{
  "compact": {
    "limits": [
      {
        "counter": "compact_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "analyze": {
    "limits": [
      {
        "counter": "analyze_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Project Graph — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON PROJECT GRAPH →

Instant setup, no code required.

FULL CATALOGUE

All 19 Project Graph tools

WRITE 5 tools

READ 14 tools

Read analyze Code quality analysis. Actions: dead_code|similar_functions|complexity|large_files|outdated_patterns|full_anal Read check_custom_rules Run custom rules analysis on a directory. Returns violations found. Read db Database analysis. Actions: schema|table_usage|dead_tables Read get_agent_instructions Get coding guidelines, architectural standards, and JSDoc rules for this project. Read get_ai_context Boot AI agent context. Two modes:\n1. Default: returns skeleton + docs (2-3k tokens) for structure overview.\n Read get_custom_rules List all custom code analysis rules. Rules are stored in JSON files in rules/ directory. Read get_focus_zone Get enriched context for recently modified files. Auto-detects from git or accepts explicit file list. Read get_framework_reference Get framework-specific AI reference documentation. Auto-detects framework from project or accepts explicit nam Read get_skeleton Get compact minified project overview (10-50x smaller than source). Returns legend, stats, node summaries, fil Read get_usage_guide Get the comprehensive usage guide for project-graph with examples and best practices.\nCall this FIRST when pl Read graph_metadata Read and validate project-local graph metadata sidecar. Actions: get|validate Read invalidate_cache Invalidate the cached graph. Use after making code changes. Read jsdoc JSDoc operations. Actions: check_consistency|check_types|generate Read navigate Navigate the project graph. Actions: expand|deps|usages|call_chain|sub_projects. deps returns code calls plus

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Project Graph

How do I prevent bulk modifications through Project Graph? +

The Project Graph server has 5 write tools including compact, docs, filters. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Project Graph.

How many tools does the Project Graph MCP server expose? +

19 tools across 4 categories: Destructive, Execute, Read, Write. 14 are read-only. 5 can modify, create, or delete data.

How do I enforce a policy on Project Graph? +

Register the Project Graph MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Project Graph tool call.

Deterministic rules across all 19 Project Graph tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON PROJECT GRAPH →

Instant setup, no code required.

19 Project Graph tools catalogued and risk-classified — across an index of 43,000+ MCP servers.