Tools / Decisions

propose_learning()

Key

Typed learning proposals from a scored outcome: an edge-confidence update, a belief supersession, a lesson, or a policy note. Everything lands at status=proposed — judgment compounds only through human approval.

propose_learning(decision_id, type, content, evidence_ids)→ PROPOSED LEARNING

Callable over MCP as propose_learning(decision_id, type, content, evidence_ids). REST mirror POST /v1/learnings Planned — alpha is MCP-only

Why your agent calls it

Where variance becomes judgment. After an outcome lands, the agent proposes what should change — an edge confidence, a superseded belief, a lesson, a policy note. Humans approve what compounds. This is the difference between a system that logs mistakes and one that stops repeating them.

What it returns

propose_learning() · Queue what the variance taught — humans approve.ILLUSTRATIVE

Call it from an agent

Over MCP, your agent invokes it on its own when the moment calls for it — just describe the situation. A Python SDK for headless runtimes is planned; this is what the call will look like.

agent · python sdkSDK · PLANNED — ALPHA IS MCP-ONLY

from cascadian import Cascadian

cx = Cascadian(api_key=os.environ["CASCADIAN_API_KEY"])

result = cx.propose_learning(
    decision_id="dec_0142",
    type="lesson",
    content="Discount pulls deals forward; net win-rate lift unclear.",
    evidence_ids=["mem_0871"],
)

Parameters

Param	Type	Description
decision_id	string	The decision whose variance prompted this.
typeREQUIRED	enum	edge_confidence · belief_supersession · lesson · policy.
contentREQUIRED	string	The proposed update, plainly stated.
evidence_ids	string[]	Outcomes or memory supporting it.

Example response

propose_learning()JSON · ILLUSTRATIVE

{
  "learning_id": "lrn_0033",
  "type": "edge_confidence",
  "target": "discount → revenue",
  "proposal": "lower confidence 0.8 → 0.6 — revenue overshot the interval twice",
  "status": "proposed"
}

Related — Decisions

keyopen_decision()Open a decision and pin what is knowable now.keyrecord_recommendation()The full recommendation — expectation frozen.scopedrecord_decision()The immutable commit. Approval enforced server-side.keyrecord_outcome()Measured actuals against the frozen expectation.keyreview_calibration()How good are the calls, over time and by type.keyget_decision()The full decision card, by id.