Extended Thinking Architect

> Category: Engineering > Domain: AI Engineering

Overview

The Extended Thinking Architect skill helps you decide when an LLM task should spend a reasoning/thinking budget, how much (no-thinking / low / medium / high), and when the better move is a cheaper model with a sharper prompt instead. It turns task signals — error cost, ambiguity, step count, latency budget — into a deterministic recommendation with a rough cost multiplier, and allocates effort across the phases of an agent loop so you front-load reasoning where it pays and avoid runaway budgets.

Clarify First

Before recommending an effort level, confirm these inputs. If any is unknown or vague, ASK — do not assume:

• ] Task type & verifiability — what the model is actually doing (extraction, classification, planning, code-debug, math…) and whether the output is checkable (sets --task-type and --verifiable)
• ] Cost of a wrong answer — how expensive a bad output is, plus the latency budget the task must fit (sets --error-cost and --latency-budget)
• ] Shape of the work — how many reasoning/tool steps are expected and how ambiguous the request is (sets --steps and --ambiguity)

Stop rule: ask only the 2-3 that most change the output. If the user says "just draft it," proceed and list your assumptions at the top of the artifact.

Quick Start

bash
# Recommend a reasoning effort level for a single task
python scripts/reasoning_budget_advisor.py --task-type code-debug \
  --error-cost high --steps 4 --ambiguity low --latency-budget interactive

# A cheap, high-volume classification task — expect "cheaper model + better prompt"
python scripts/reasoning_budget_advisor.py --task-type classification \
  --error-cost low --latency-budget realtime --json

# Allocate reasoning effort across the phases of an agent loop
python scripts/reasoning_loop_allocator.py --difficulty high --steps 8 \
  --max-budget-multiplier 30

# Tight-latency loop — see effort capped per phase
python scripts/reasoning_loop_allocator.py --difficulty medium --steps 5 --realtime --json

Tools Overview

| Tool | Purpose | Key Flags | |------|---------|-----------| | reasoning_budget_advisor.py | Recommend an effort level (none/low/medium/high) or "prompt-first / cheaper-model" for one task, with rationale + cost multiplier | --task-type, --error-cost, --steps, --ambiguity, --latency-budget, --verifiable, --json | | reasoning_loop_allocator.py | Allocate reasoning effort across agent-loop phases (plan/act/observe/recover/finalize) under a total budget cap | --difficulty, --steps, --max-budget-multiplier, --realtime, --json |

Workflows

Choosing Effort for a New Task

1. Identify the task type and whether the output is verifiable (ground truth or a checker exists).
2. Run reasoning_budget_advisor.py with the error cost, step count, ambiguity, and latency budget.
3. If the result is prompt-first, fix the prompt/spec (clarify, add examples) before spending any reasoning, then re-run.
4. If the result is cheaper-model, route to a smaller/faster model and invest the savings in a better prompt.
5. Otherwise adopt the recommended effort, note the cost multiplier, and set a per-call budget cap.

Budgeting Reasoning Across an Agent Loop

1. Estimate overall task difficulty and the expected number of steps.
2. Run reasoning_loop_allocator.py to get per-phase effort (front-loaded at plan/recover, thin at act/observe).
3. Apply the total budget cap as a hard stop so a stuck loop cannot run away.
4. Instrument per-phase token spend; if observe/act phases consume high reasoning, that is an overthinking signal — clamp them.

Reference Documentation

• When to Use Extended Thinking - Decision matrix of task classes where reasoning pays off vs. is wasted, interaction with tool use and agent loops, budget guards, overthinking failure modes, and eval signals.
• Reasoning Budget Patterns - Allocation patterns, escalation ladders, caps and circuit breakers, and the cost/quality/latency tradeoff model.

Common Patterns

When Reasoning Pays Off

• Multi-step deduction with a verifiable answer (math, constraint solving, debugging from a stack trace)
• Planning and decomposition before a long agent run — front-load thinking once, not on every tool call
• High error-cost decisions where a wrong answer is expensive to detect or undo

When Reasoning Is Wasted

• Extraction, classification, and formatting — deterministic mappings, not deduction; a cheaper model usually wins
• Underspecified requests — extra thinking confidently elaborates on the wrong goal; fix the prompt first
• Realtime/latency-tight paths where thinking tokens blow the budget more than they improve quality

Guarding the Budget

• Set a per-call effort cap and a loop-level total cap (e.g. a multiple of one no-thinking call)
• Escalate effort only on failure (retry at higher effort), never start high "to be safe"
• Treat reasoning spent on trivial sub-steps as a regression — alert on per-phase token spend