Cost-Aware Prompting Playbook for Advanced CLI Workflows

Use this playbook when running long, multi-phase CLI delivery loops where token spend is driven more by context carryover than by output size.

Quick Routing Matrix

Work shape	Session move	Model default	Execution style
Mechanical cleanup (branch/worktree hygiene, scans, formatting)	Stay in session; /compact at phase boundary	claude-haiku-4.5 or gpt-5.4-mini	Interactive or delegated batch
Moderate reasoning (triage, planning, scoped reviews)	Stay in session if same objective; /compact on phase switch	claude-sonnet-5 (or equivalent reasoning tier)	Delegate scan work, keep decisions in main thread
Hard ambiguity (architecture, deep debugging, security tradeoffs)	Use a focused session if objective changes	claude-opus-4.8 (or top-tier equivalent)	Time-boxed deep pass, then downshift

Phase-Boundary Rule: /compact vs /new

1. Use /compact when the objective is unchanged but the phase changes: triage -> implementation -> merge waiting -> closeout.
2. Use /new when objective or deliverable changes: cleanup -> RCA, implementation -> unrelated docs pack, repo A -> repo B.
3. Before either command, preserve a canonical handoff pointer: issue URL, active PR URL, branch, and one short done/next/blocked note.

Fleet and Delegation Rule

Use delegation when the task is broad but mechanical. Keep the main thread decision-only.

• Delegate: repository scans, issue sweeps, dependency checks, repetitive status loops.
• Keep in main thread: merge decisions, release gate approvals, risk acceptance.

Batch related work in one kickoff and monitor with /tasks rather than repeated restarts of the same orchestration intent.

Durable Learnings Pattern

When a cycle finishes, capture one compact learning artifact so the next run does not require prompt reconstruction.

Recommended capture shape:

Cycle summary
- Scope: <issue/goal>
- Decisions: <2-4 bullets>
- Evidence: <PR/workflow links>
- Next trigger: <what starts next cycle>
- Stop condition met?: <yes/no + reason>

Store this where the team can reuse it (issue/PR comment, runbook section, or canonical summary file path referenced by future sessions).

Minimal Start Prompt Template

Goal: <single objective>
References: <issue/PR/spec links>
Mode: <interactive|fleet>
Constraints: required checks, merge pacing, stop condition
Cost policy: compact at phase boundaries; new session on domain pivot

Exit Checklist

• Objective completed or explicit stop condition reached.
• Required checks and evidence links captured.
• Branch/worktree cleanup status recorded.
• Next cycle can start from links, not transcript replay.