neurogenesis
Developmental agents for the agent economy: create an agent from a digital genome, then evolve it wi
Should I use this
Quality & Safety
Findings (1)
- LOWTool 'describe_agent' description lacks action verbin describe_agent
Based on automated analysis of tool definitions and protocol compliance.
Context Cost
This is the approximate number of tokens consumed each time the server's tools are loaded into a model's context. Higher counts reduce the attention available for other tasks.
Install
One-Click Install
Add this to your `claude_desktop_config.json` file:
{
"mcpServers": {
"neurogenesis": {
"url": "https://mcp.viridisconservation.com/neurogenesis/mcp"
}
}
}Remote endpoints
https://mcp.viridisconservation.com/neurogenesis/mcpstreamable-httpWhat it can do
Tool inventory
Tools (14)
🟡create_agent(genome, request_id)
Create a developmental agent from a digital genome: {agent_name, purpose, initial_nodes (unique, >=1), fitness_metrics (>=1), optional growth_rules / safety_axioms}. Returns agent_id + initial graph summary.
Input Schema
{
"type": "object",
"properties": {
"genome": {
"additionalProperties": true,
"title": "Genome",
"type": "object"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"genome"
],
"title": "create_agentArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "create_agentOutput"
}🟡submit_evaluation(agent_id, evaluation, request_id)
Evolve an agent with one task outcome: {task_id, task_type, success_score in [0,1], optional accuracy/user_satisfaction/ cost_efficiency/safety_score/notes/used_nodes/used_edges}. Success strengthens the used edges, failure weakens them; growth and pruning follow the genome's rules under its safety axioms (NG1). Returns the new developmental-ledger events.
Input Schema
{
"type": "object",
"properties": {
"agent_id": {
"title": "Agent Id",
"type": "string"
},
"evaluation": {
"additionalProperties": true,
"title": "Evaluation",
"type": "object"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"agent_id",
"evaluation"
],
"title": "submit_evaluationArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "submit_evaluationOutput"
}🟢get_agent(agent_id)
Full current state: genome, cognitive graph (nodes/edges with weights and trust), and summary counts.
Input Schema
{
"type": "object",
"properties": {
"agent_id": {
"title": "Agent Id",
"type": "string"
}
},
"required": [
"agent_id"
],
"title": "get_agentArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "get_agentOutput"
}🟢list_agents
All developmental agents on this mount, with summary counts.
Input Schema
{
"type": "object",
"properties": {},
"title": "list_agentsArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "list_agentsOutput"
}⚪best_next_steps(agent_id, from_node, limit)
Routing recommendation: the strongest next cognitive steps from a given node, by learned edge weight and trust.
Input Schema
{
"type": "object",
"properties": {
"agent_id": {
"title": "Agent Id",
"type": "string"
},
"from_node": {
"title": "From Node",
"type": "string"
},
"limit": {
"default": 3,
"title": "Limit",
"type": "integer"
}
},
"required": [
"agent_id",
"from_node"
],
"title": "best_next_stepsArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "best_next_stepsOutput"
}🟢get_ledger(agent_id, limit)
The append-only developmental ledger: every growth, pruning, and evaluation event with reasons (NG3 — returned verbatim).
Input Schema
{
"type": "object",
"properties": {
"agent_id": {
"title": "Agent Id",
"type": "string"
},
"limit": {
"default": 100,
"title": "Limit",
"type": "integer"
}
},
"required": [
"agent_id"
],
"title": "get_ledgerArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "get_ledgerOutput"
}⚪export_state(agent_id)
Portable state document for an agent (import_state recreates it anywhere — including a self-hosted verdigraph-neurogenesis).
Input Schema
{
"type": "object",
"properties": {
"agent_id": {
"title": "Agent Id",
"type": "string"
}
},
"required": [
"agent_id"
],
"title": "export_stateArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "export_stateOutput"
}⚪import_state(state, request_id)
Recreate an agent from an export_state document.
Input Schema
{
"type": "object",
"properties": {
"state": {
"additionalProperties": true,
"title": "State",
"type": "object"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"state"
],
"title": "import_stateArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "import_stateOutput"
}🔴delete_agent(agent_id, request_id)
Remove a developmental agent from this mount.
Input Schema
{
"type": "object",
"properties": {
"agent_id": {
"title": "Agent Id",
"type": "string"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"agent_id"
],
"title": "delete_agentArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "delete_agentOutput"
}🟢register_compute_profile(profile, request_id)
Register a caller-owned Wu Wei execution profile: reuse/cache, deterministic rule, local model, tool/workflow, or cloud model. Profiles can declare quality, reliability, capabilities, token costs, latency, locality, energy rates or average power, carbon intensity, and cache confidence/age. Routing never invents capacity or energy evidence.
Input Schema
{
"type": "object",
"properties": {
"profile": {
"additionalProperties": true,
"title": "Profile",
"type": "object"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"profile"
],
"title": "register_compute_profileArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "register_compute_profileOutput"
}⚪route_task(task, request_id)
Choose the least-burden eligible route for a task. Hard constraints include quality, reliability, locality, capabilities, context, cost, latency, and energy. Optional baselines quantify predicted savings. Explicit allow_defer/value/urgency fields may produce a no-work decision; no result is then claimed. Returns a hash-bound decision receipt.
Input Schema
{
"type": "object",
"properties": {
"task": {
"additionalProperties": true,
"title": "Task",
"type": "object"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"task"
],
"title": "route_taskArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "route_taskOutput"
}⚪record_route_outcome(decision_id, success_score, actual_cost_usd, actual_latency_ms, actual_energy_wh, ...)
Attach one observed outcome to a Wu Wei decision. Actual cost, latency, and energy are optional and remain explicitly unknown when omitted. One append-only, hash-bound outcome is allowed per decision.
Input Schema
{
"type": "object",
"properties": {
"decision_id": {
"title": "Decision Id",
"type": "string"
},
"success_score": {
"title": "Success Score",
"type": "number"
},
"actual_cost_usd": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"title": "Actual Cost Usd"
},
"actual_latency_ms": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"title": "Actual Latency Ms"
},
"actual_energy_wh": {
"anyOf": [
{
"type": "number"
},
{
"type": "null"
}
],
"default": null,
"title": "Actual Energy Wh"
},
"notes": {
"default": "",
"title": "Notes",
"type": "string"
},
"request_id": {
"anyOf": [
{
"type": "string"
},
{
"type": "null"
}
],
"default": null,
"title": "Request Id"
}
},
"required": [
"decision_id",
"success_score"
],
"title": "record_route_outcomeArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "record_route_outcomeOutput"
}🟢compute_efficiency_report(limit)
Free read: decision/outcome receipts, route modes, compute avoided, predicted savings, predicted-vs-observed cost/latency/energy coverage, and Landauer-floor context. Estimates and observations stay distinct.
Input Schema
{
"type": "object",
"properties": {
"limit": {
"default": 50,
"title": "Limit",
"type": "integer"
}
},
"title": "compute_efficiency_reportArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "compute_efficiency_reportOutput"
}🟢describe_agent
Return capabilities and input contract.
Input Schema
{
"type": "object",
"properties": {},
"title": "describe_agentArguments"
}Output Schema
{
"type": "object",
"properties": {
"result": {
"title": "Result",
"type": "string"
}
},
"required": [
"result"
],
"title": "describe_agentOutput"
}Community
Evidence