# Chapter 7 lab — the planning paradigms

Three recipes covering planner, rewoo, and blackboard, plus the captured governance catch.

# Files

• plan_repair.erpe — planner: plan → verify → summarize, no steps (the plan document is the product; modeled on the shipped plan.erpe).
• rewoo_outline.erpe — rewoo: required plan + synthesize, no steps.
• board_survey.erpe — blackboard: two authored sources (read [state.x] lanes, mandatory write), no do legs — the data-flow skeleton form.

# Reproduce

relurplint --check recipes --workspace .   # 0 diagnostics (first pass)
# restart relurpish, pin, dispatch:
#   /recipe euclo.thoughtrecipe.plan_repair

# Captured: governance catches a malformed tool call

The pinned plan_repair dispatch on the verification machine:

{"decided_by": "pinned", "node.id": "euclo.dispatch"}
{"message": "node planner_execute execution failed: tool file_read blocked: input schema invalid: $.path required"}
{"outcome": "failure", "llm_calls": "0",
 "outcome_kind": "execution failed: node planner_execute execution failed: tool file_read blocked: input schema invalid: $.path required"}

The planning model omitted file_read’s required path argument; the tool-manifest schema validation blocked the call at execution, named the JSON path, and the default step-error policy aborted the route. Fix: /retry, or a model that emits complete arguments.

# Flagged, not explained

llm_calls reads 0 for planner and chainer runs while react turns counted (Chapter 2: five). Pattern suggests the counter is react-bound as of the pin. [NEEDS CONFIRMATION: llm_calls counting for non-react paradigms at your pin]

# Contract pointers

• planner/rewoo/htn/blackboard: cognitionzoo/<paradigm>/contract.go
• htn composes react; blackboard authored sources: cognitionzoo/blackboard/authored_source.go (“authored structure is authoritative”; no model calls beyond a source’s own execution)
• do literal-input limitation: recipe DSL manual §8.8