io.github.JaysonAIOnline/cognitive-substrate

mcp-cognitive-substrate

mcp-name: io.github.JaysonAIOnline/cognitive-substrate

A 28-layer cognitive substrate with cross-session Tree-of-Thoughts (ToT) evolutionary memory, conditional self-telemetry, and A2A tools for MCP agents.

Lets agents reason through a validated 28-layer substrate, evolve memory across sessions, and communicate with peer agents — all in one pip-installable package.

Features

• 28-layer cognitive substrate with Pydantic validation — CognitiveSubstrate validates reasoning through 6 families / 28 layers.
• Cross-session ToT evolutionary memory — SQLite-backed tree-of-thoughts nodes + substrate history; pruned branches become lessons for future sessions.
• Robust stack-based JSON parser — no regex; handles nested brackets, escaped strings, embedded code fences (robust_slice / robust_json_slice).
• Self-telemetry tool — get_cognitive_tree_state returns active paths and pruned branches for a session.
• Post-execution storage loop — store_5key_telemetry auto-saves compact 5-key telemetry (foundations, metacognition, defensive, resource, utility).
• 7 reasoning paradigms — deductive, inductive, abductive, analogical, causal, syllogistic, falsification.
• A2A tools — list, discover, call, and orchestrate peer agents.
• MCP server — exposes everything as tools via the cognitive-substrate CLI.

Install

pip install mcp-cognitive-substrate

Or install from source:

git clone https://github.com/JaysonAIOnline/mcp-cognitive-substrate.git
cd mcp-cognitive-substrate
pip install -e .[test]

Requires Python >= 3.11.

Quick Start

from mcp_cognitive_substrate.substrate import CognitiveSubstrate
from mcp_cognitive_substrate.memory import get_cognitive_tree_state, store_5key_telemetry

substrate = CognitiveSubstrate()
response = substrate.run("Your user prompt here")
print(response["layers_applied"], "layers applied")
print(response["substrate_verdict"])

Usage

28-layer substrate

from mcp_cognitive_substrate import substrate

# Layer count and schema
print(substrate.layer_count())        # 28
print(substrate.SUBSTRATE_SCHEMA)     # the full 6-family schema

# Validate a prompt through the substrate
result = substrate.CognitiveSubstrate(session_id="s1").run("deploy safely")
print(result["substrate_verdict"])    # heuristic pruning verdict

# Run a single paradigm
from mcp_cognitive_substrate import run_paradigm
print(run_paradigm("14_idempotency_side_effect_audit", {"evaluate_branch": True}))

Cross-session ToT evolutionary memory

from mcp_cognitive_substrate.memory import (
    store_5key_telemetry,
    get_cognitive_tree_state,
    prune_failed_approach,
)

node_id = store_5key_telemetry(
    session_id="session-a",
    payload={
        "foundations": {"premise_validation": "assuming deps", "state_hash": "h", "falsification_notes": "deps missing"},
        "defensive": {"blast_radius": "unpredictable", "is_idempotent": True, "invariant_rule": "r"},
        "resource": {"big_o": "o(n)", "latency_bottleneck": "none"},
        "utility": {"load_summary": "pin versions to deploy", "checklist_verified": True},
        "metacognition": {"self_critique": "c", "drift_pct": 0.1},
    },
    score_delta=-110.0,
)
prune_failed_approach(node_id)
state = get_cognitive_tree_state("session-a", include_pruned=True)
print(state["active_path_count"], state["pruned_branch_count"])

Stack-based JSON parser

from mcp_cognitive_substrate.memory import robust_slice, robust_json_slice

cleaned, payload = robust_slice('prefix {"a": {"b": [1, 2]}, "c": "x"} suffix')
# payload == {"a": {"b": [1, 2]}, "c": "x"}; cleaned == "prefix suffix"

7 reasoning paradigms + A2A

from mcp_cognitive_substrate import reason, a2a_list, a2a_call, a2a_orchestrate

print(reason("Solve X", reasoning_type="abductive", depth=3)["steps"])
print(a2a_list())
print(a2a_call("peer-agent", "hello"))
print(a2a_orchestrate("hi", capability="memory"))

As an MCP server

cognitive-substrate          # starts stdio MCP server
cognitive-substrate --info   # prints package summary

All of the above — substrate paradigms, extraction/evaluation, memory store/recall, ToT lessons, tree-state telemetry, JSON parsing, reasoning plans, and A2A — are exposed as MCP tools.

Testing

pip install -e .[test]
python -m pytest src/tests -q     # 16 tests

License

MIT