optimize git:20261030.7d3bf2cA

Immutable. This exact content never changes and is served at /api/v1/blob/f73213e1dcbfa874.

---
name: optimize
description: API cost & performance scanner — identify expensive operations and optimization opportunities
argument-hint: Optional focus area (e.g., "ai", "caching", "queries", "latency")
---

# /optimize — API Cost & Performance Scanner

You are scanning this project's API and infrastructure for cost and performance optimization opportunities.

## Ethos

!`test -s .adlc/ETHOS.md && cat .adlc/ETHOS.md || echo No ethos found — run /init, or /adlc:init on the plugin install, to vendor .adlc/ETHOS.md`

## Context

Run this block first, with your shell tool, and read its output before you continue. Each read prints a notice instead of failing when its file is absent.

```sh
# adlc-context
echo '--- Architecture'
cat .adlc/context/architecture.md 2>/dev/null || echo 'No architecture context found'
echo '--- Project overview'
cat .adlc/context/project-overview.md 2>/dev/null || echo 'No project overview found'
```

## Input

The focus is the argument: whatever follows `/optimize` in the invocation (on Claude Code it can also arrive as an `ARGUMENTS:` line appended to this text), a dimension or an area. With none, Step 1 scans all dimensions.

## Prerequisites

Before proceeding, verify that `.adlc/context/architecture.md` and `.adlc/context/project-overview.md` exist. If any of these files are missing, stop and tell the user: "The `.adlc/` structure hasn't been initialized. Run `/init` (`/adlc:init` on the plugin install) first to set up the project context."

## Instructions

### Step 1: Determine Focus
1. If given a focus area, prioritize that dimension
2. If no argument, scan all dimensions
3. Read `.adlc/context/architecture.md` for current caching and optimization patterns

### Step 2: Launch Scanner Agents
Launch 3 formal scanner agents in parallel using the Agent tool. Each agent is defined in `$ADLC_TOOLKIT_ROOT/agents/` with its full scanning checklist and model selection (sonnet for deep analysis).

If this harness can run subagents, dispatch these agents in parallel; if it cannot, or you are a subagent, run each checklist inline from `.adlc/agents/<name>.md`. Agent names are bare on the symlink install and `adlc:`-prefixed on the plugin install (`adlc:<name>`): before dispatching, check which form the Agent tool lists and use that one, and do not wait for an error, because a name it does not list can silently run a generic agent without the checklist; if neither form is listed, say which agent is missing and run its checklist inline. When `.adlc/agents/` is absent, say that `/init` (`/adlc:init` on the plugin install) vendors it, then read `$ADLC_TOOLKIT_ROOT/agents/<name>.md` (the toolkit root is the parent of this skill's base directory) where your file tool can reach it; never improvise a checklist. An inline run of a review or audit agent is read-only: report findings, apply no fixes, and run no test runner, build tool, package manager or interpreter.

1. **api-cost-scanner** agent — provide the focus scope and architecture.md for caching context
2. **db-perf-scanner** agent — provide the focus scope
3. **latency-scanner** agent — provide the focus scope

Each agent returns structured findings with estimated impact, effort, and risk.

### Step 3: Build Optimization Report

#### Cost Summary
| Service | Est. Monthly Usage | Est. Monthly Cost | Top Optimization |
|---------|-------------------|-------------------|------------------|
| Claude Sonnet | X calls | $X | ... |
| Claude Haiku | X calls | $X | ... |
| Gemini Flash | X calls | $X | ... |
| SerpAPI | X calls | $X | ... |
| Firestore | X reads/writes | $X | ... |
| GCS | X operations | $X | ... |

#### Performance Hotspots
Rank endpoints by estimated latency, highlighting:
- Sequential operations that could be parallel
- Missing caches
- N+1 query patterns
- Unnecessary data fetching

#### Optimization Opportunities
For each opportunity:
- **What**: Description of the optimization
- **Impact**: Cost savings or latency reduction estimate
- **Effort**: Small / Medium / Large
- **Risk**: Low / Medium / High (chance of breaking something)

### Step 4: Prioritized Recommendations
1. Rank optimizations by impact/effort ratio
2. Group into quick wins (small effort, high impact) and strategic improvements
3. Suggest which items warrant a full ADLC requirement (candidates for `/spec`)