Server data from the Official MCP Registry
Paid hosted MCP for agent-to-agent compute routing. 25% of net revenue funds conservation.
Paid hosted MCP for agent-to-agent compute routing. 25% of net revenue funds conservation.
Remote endpoints: streamable-http: https://verdigraph-mcp.hartjustin6.workers.dev/mcp sse: https://verdigraph-mcp.hartjustin6.workers.dev/mcp
Verdigraph is a sophisticated MCP server framework for self-evolving AI agents with a hosted paid component. The codebase is well-structured with proper authentication (OAuth 2.1 + PKCE on hosted version), input validation via Pydantic/Zod schemas, and reasonable permission scoping. However, several moderate concerns exist: the open-source stdlib version lacks authentication entirely, file I/O permissions are broadly scoped without path validation, the hosted version has undisclosed dependency risks in the Node.js stack, and sensitive billing/credential handling in the hosted component requires careful auditing. These issues are mitigated by the fact that the stdio server is intended for local use and the hosted version enforces OAuth, but users should understand the permission model. Supply chain analysis found 10 known vulnerabilities in dependencies (1 critical, 4 high severity).
6 files analyzed · 19 issues found
Security scores are indicators to help you make informed decisions, not guarantees. Always review permissions before connecting any MCP server.
This plugin requests these system permissions. Most are normal for its category.
Available as Local & Remote
This plugin can run on your machine or connect to a hosted endpoint. during install.
From the project's GitHub README.
Clone this repo, run one script, and within 60 seconds you're building deterministic, content-addressed brain artifacts from any agent file — Claude project export, OpenAI Assistant config, raw prompt list, or Verdigraph genome JSON. Pure Python core; zero external services required.
git clone https://github.com/viridis-security/verdigraph-neurogenesis
cd verdigraph-neurogenesis
bash quickstart.sh
That's it. The script creates a venv, installs the package editable, runs the brain builder against an example genome, and prints the deterministic brain_id + content_hash. No Cloudflare account, no Stripe key, no verdigraph.dev account needed. Everything runs locally.
If you also have an internet connection, the script will additionally hit https://verdigraph.dev/app/import with the same input bytes and confirm the hosted Worker produces the exact same brain_id — that's your proof the local build is byte-equivalent to the production reference implementation.
Verdigraph turns an agent file into an inspectable cognitive graph with a content-addressed identifier you can pin in git, cite in an audit, or paste into a code review. Three things make this useful:
brain_id, content_hash, and graph structure. Run it twice, get the same answer twice. Run it in Python locally; run it in TypeScript on the Worker; same answer either way.I9_fitness_metric_wired) so you can prove what the agent file actually compiles to without trusting a black box.verdigraph/brain.py (≈ 660 lines) in an afternoon.python -m verdigraph build --file examples/hypothetical_research_agent.genome.json --format verdigraph_genome --pretty
Or pipe input:
cat my_agent.json | python -m verdigraph build --stdin --format auto --summary --pretty
python -m verdigraph build --file my_claude_project_export.json --format claude_project_export --pretty
python -m verdigraph build --file my_assistant.json --format openai_assistant --pretty
echo -e "You are a helpful assistant.\nSummarize the user's request.\nPlan steps and execute." \
| python -m verdigraph build --stdin --format prompt_list --pretty
python -m verdigraph build --file my_agent.json --pretty > brain.json
python -m verdigraph verify brain.json
from verdigraph.brain import extract, verify_brain, to_dict
genome = b'{"agent_name":"my_agent","purpose":"...","initial_nodes":["planner","executor"],"fitness_metrics":["task_success_rate"]}'
brain = extract("verdigraph_genome", genome)
print(brain.brain_id) # e.g. RMX124YY916WP0TCSEHFYX7M30
print(brain.brain_uri) # verdigraph://brain/RMX124YY916WP0TCSEHFYX7M30
print(brain.content_hash) # sha256 hex
print(len(brain.nodes), "nodes,", len(brain.edges), "edges")
report = verify_brain(brain)
assert report.passed # all non-advisory invariants pass
print(to_dict(brain)) # serialize for storage / round-trip
pip install -e ".[mcp]"
verdigraph-mcp # runs over stdio
Then in Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"verdigraph": {
"command": "/absolute/path/to/repo/.venv/bin/verdigraph-mcp",
"args": []
}
}
}
Or in Claude Code: claude mcp add --transport stdio verdigraph /absolute/path/to/repo/.venv/bin/verdigraph-mcp.
Restart your client. Your agent now has verdigraph_* tools to build/verify/evolve brains directly. No network calls; everything runs on your machine.
| Field | What it is | How to verify |
|---|---|---|
brain_id | 26-char Crockford-base32; derived from sha256(input_bytes + b":" + format) | python -m verdigraph build --file <same bytes> — same id every time |
brain_uri | verdigraph://brain/<brain_id> | Self-describing form; safe for content-safety classifiers |
content_hash | sha256(canonicalize(brain_body_minus_content_hash)) | See docs/CANONICALIZATION.md for the exact algorithm |
input_sha256 | sha256(raw_input_bytes) | sha256sum your_file.json |
| Invariant report | 9 required checks + 1 advisory I9_fitness_metric_wired | All carry id, description, passed, optional passed_with_default, advisory, detail |
Apply json.dumps with separators=(",", ":") after recursively sorting every object's keys lexicographically by codepoint and coercing integer-valued floats to integers (matches JavaScript JSON.stringify byte-for-byte). UTF-8 encoded before hashing. See verdigraph/brain.py::canonicalize (≈ 20 lines, stdlib only).
verdigraph-neurogenesis/
├── README.md ← you are here
├── quickstart.sh ← clone → first brain in 60 seconds
├── pyproject.toml ← Python package metadata
├── verdigraph/ ← Python core (no external deps)
│ ├── brain.py ← deterministic build pipeline (extract / canonicalize / verify / evolve)
│ ├── cli.py ← `python -m verdigraph` CLI
│ ├── genome.py ← AgentGenome / GrowthRules / SafetyAxioms (live-agent runtime)
│ ├── graph.py ← CognitiveGraph / CognitiveNode / CognitiveEdge
│ ├── agent.py ← DevelopmentalAgent (live-agent runtime)
│ ├── growth.py / pruning.py ← evolution operators
│ ├── evaluation.py ← task-outcome ledger
│ ├── compute.py ← compute-routing helpers
│ └── ledger.py ← immutable event log
├── verdigraph_mcp/ ← optional: stdio MCP server (`pip install -e ".[mcp]"`)
├── tests/ ← pytest, all green on a clean clone
├── examples/ ← runnable demos with fixture genomes
├── docs/ ← canonicalization spec, architecture, invariants
├── papers/ ← three companion papers (Zenodo-archived)
└── hosted-mcp/ ← OPTIONAL: Cloudflare Workers deployment if you want a hosted instance
A reference Cloudflare Workers deployment lives in hosted-mcp/. It serves the same deterministic-build pipeline over HTTPS + OAuth 2.1 + PKCE, adds prepaid USD credits via Stripe, and Ed25519-signed compliance attestations. You do not need this to use the Python core. It exists because the same protocol can run hosted if you want a shared multi-caller environment. See hosted-mcp/README.md for deployment instructions.
A live reference deployment runs at https://verdigraph.dev — same byte-equivalent pipeline. The local Python implementation is the canonical source; the Worker is a reimplementation for hosting convenience.
Python core:
source .venv/bin/activate
pip install -e ".[dev]"
pytest -q
TypeScript hosted-MCP (Cloudflare Worker):
cd hosted-mcp
npm ci
npm run typecheck
npm test
Both suites run in CI (.github/workflows/tests.yml) on every push and pull
request: the Python job across 3.10 / 3.11 / 3.12, and the hosted-mcp job on
Node 22 — where the cross-core parity.test.ts executes against a real Python
install rather than self-skipping. A secret-scan job fails the build if a live
Stripe identifier is ever committed.
The tests/test_brain_parity.py suite locks the deterministic-build contract — specifically that b'{"agent_name":"x","purpose":"y","initial_nodes":["a"],"fitness_metrics":["task_success_rate"]}' produces brain_id == "RMX124YY916WP0TCSEHFYX7M30" and content_hash == "20b9e5be0e5a0d34e564df6d0a554b1232ff9cc3ff309ab8da77a97756602c0c". If either side ever drifts, that test fails on the next CI run and we ship the divergence as a deliberate schema bump.
In papers/:
PAPER_1_Physical_NeuroGenesis_SynapseForge.md — physical version: AI-agent-architected, 3D-printed, solution-grown neuromorphic substrates.PAPER_2_Verdigraph_Digital_NeuroGenesis.md — software version: self-evolving digital cognitive graphs.PAPER_3_Verdigraph_Compute_Efficiency.md — compute-efficiency layer.To cite:
Hart, Justin. (2026). Verdigraph NeuroGenesis: A Software Framework for Self-Evolving AI-Agent Cognitive Substrates (Version 0.1.0). Zenodo. https://doi.org/10.5281/zenodo.20261687
MIT. Maintained by Viridis LLC. Contact: hartjustin6@gmail.com.
This is an experimental research framework. It does not create autonomous unrestricted self-modifying AI. All growth and pruning actions are constrained by explicit genome rules, safety invariants, and an auditable ledger.
Be the first to review this server!
by Modelcontextprotocol · Developer Tools
Web content fetching and conversion for efficient LLM usage
by Modelcontextprotocol · Developer Tools
Read, search, and manipulate Git repositories programmatically
by Toleno · Developer Tools
Toleno Network MCP Server — Manage your Toleno mining account with Claude AI using natural language.