Proposed specificationSource-located exemplarH1 H3

Dual-loop assistance

Separate planning and execution feedback.

JOIN POINTplanner_executor_feedback_linkRECORD STATUSCandidate · v0.1.0

UML class view

The proposed contract drawn as participant roles, received calls, and numbered dependencies.

Object interactions

The same roles and calls as a class collaboration and a numbered message sequence.

Object interaction diagram for Dual-loop assistance, showing role collaboration and a numbered message sequence — FIG. E09 / OBJECT INTERACTIONS
Open full SVG Download PDF

Pointcut

Where the mechanism applies. All four dimensions belong to the proposed match, not just its vocabulary.

Context: A planner or assistance loop and an executor loop operate concurrently or iteratively.
Trigger: Executor state or environment feedback updates planning.
Structure: Planner, executor, feedback channel, synchronization, and authorization boundary.
Effect: The next plan or assistance action changes in response to attributable executor feedback.

Protected contract

The obligation that clients, operators, and auditors should be able to rely on.

Requires: Planner and executor states have compatible identities and timing.
Ensures: Feedback reaches planning before the affected action when required.
Invariant: The outer loop cannot bypass executor authorization for physical or external actions.
Violation: Stale, conflicting, or unavailable feedback has a safe coordination policy.

Negative control

Two unrelated loops without a feedback link do not meet the pointcut.

Variation & lifecycle

The hot spot is admitted only where the protected semantics survive.

Permitted variation

Reactor-assisted executor, dual loop, planner-executor, and assistance policy.

Test obligation

Test feedback timing, stale feedback, authorization, and loop termination.

Evolution question

Did assistance move from periodic prompting to a synchronized outer loop?

Prompt + code design

An illustrative prompt and a small design sketch for the proposed contract. Names and APIs are invented; this is not a repository excerpt.

PROMPT SKETCH

Link planner and executor state with compatible identities and timing. Deliver corrective feedback before the affected action when the reactor's deadline requires it.

CODE DESIGN / PYTHON

def assist(plan, execution, reactor):
    for step in execution.pending():
        hint = reactor.observe(plan.state, step)
        if hint and hint.deadline <= step.start:
            plan = plan.with_hint(hint)
            execution = execution.revise(step, hint)
    return execution.finish(plan)

Illustrative design sketch. Assistance is a timing contract; late feedback is a postmortem, not a correction. This example is a teaching aid, not implementation evidence or a conformance claim.

Detection & matching

The frozen rule that retrieves this record, the canonical source exemplar, and the control that should stay unmatched.

Join point: planner_executor_feedback_link
Source locator: See the source record.
Lexical cue: (?i)\b(?:dual[_-]?loop|planner[_-]?executor|assistance[_-]?loop|planning[_-]?feedback)\bFrozen expression, v0.1.0. A match is a retrieval cue, not a confirmed implementation.
Aliases: dualloop, plannerexecutor, assistanceloop, planningfeedback
Canonical exemplar: JIA-Lab-research/CarryOnsrc/carryon/main.py:176Support: README.md:43
Negative control: Two unrelated loops without a feedback link do not meet the pointcut.

Scope-anchored review. pohe-agent analyze <repo> --pattern E09 anchors the cue to its enclosing declaration and checks a separate operational signature for all four pointcut dimensions. Missing roles and negative-control risks remain visible; a structural match is not E2/E3. Calibration report ↗

Repository files

Up to five public files retrieved by the frozen cue and pinned to the audited commit. A match is a place to inspect, not a confirmed implementation.

01JIA-Lab-research/CarryOnsrc/carryon/main.py:176CANONICAL SOURCE 02crewAIInc/crewAI

lib/crewai/tests/cassettes/agents/TestResponseFormatWithKickoff.test_kickoff_response_format_with_planning_and_tools.yaml:1222

RETRIEVAL CUE 03mingju-c/HarnessForgeHarnessForge_4B/harness_factory/rounds/round_00_base/harness3/Description.md:8RETRIEVAL CUE

Evidence & provenance

Each layer answers a different question. Do not substitute retrieval for review.

Source-located exemplar

A concrete source anchor is available for this proposed record. It is one exemplar, not evidence that all repositories—or every contract field—conform.

Review boundary: this candidate and its specification come from an AI-assisted study. Independent human validation remains open.

ANY-EVIDENCE FAMILY COVERAGE3/36

8.3% of repository families in the frozen panel have at least one current cue, historical cue, or canonical source layer. This is not confirmed prevalence.

Current lexical-cue repositories: 3 / 41

Canonical source anchorJIA-Lab-research/CarryOn 02c97b80987csrc/carryon/main.py:176

Supporting surfaceREADME.md:43

Reactor closes event-feedback-planning-dispatch loop alongside execution.

Exploratory retrieval cue Regex / v0.1.0

(?i)\b(?:dual[_-]?loop|planner[_-]?executor|assistance[_-]?loop|planning[_-]?feedback)\b

This expression is a discovery instrument only. A match is a retrieval cue—not a confirmed implementation, pointcut match, or contract proof.