agent-auditAudit your AI agent setup for performance, cost, and ROI. Scans OpenClaw config, cron jobs, session history, and model usage to find waste and recommend opti...
Install via ClawdBot CLI:
clawdbot install sharbelayy/agent-auditScan your entire OpenClaw setup and get actionable cost/performance recommendations.
python3 {baseDir}/scripts/audit.pyOptions:
python3 {baseDir}/scripts/audit.py --format markdown # Full report (default)
python3 {baseDir}/scripts/audit.py --format summary # Quick summary only
python3 {baseDir}/scripts/audit.py --dry-run # Show what would be analyzed
python3 {baseDir}/scripts/audit.py --output /path/to/report.md # Save to file~/.openclaw/openclaw.json or similar)Classify each task into complexity tiers:
| Tier | Examples | Recommended Models |
|------|----------|-------------------|
| Simple | Health checks, status reports, reminders, notifications | Cheapest tier (Haiku, GPT-4o-mini, Flash, Grok-mini) |
| Medium | Content drafts, research, summarization, data analysis | Mid tier (Sonnet, GPT-4o, Pro, Grok) |
| Complex | Coding, architecture, security review, nuanced writing | Top tier (Opus, GPT-4.5, Ultra, Grok-2) |
Classification signals:
For each task where the model tier doesn't match complexity:
ā ļø RECOMMENDATION: Downgrade "Knox Bot Health Check" from opus to haiku
Current: anthropic/claude-opus-4 ($15/M input, $75/M output)
Suggested: anthropic/claude-haiku ($0.25/M input, $1.25/M output)
Reason: Simple status check averaging 300 output tokens
Estimated savings: $X.XX/month
Risk: LOW ā task is simple pattern matching
Confidence: HIGHOutput a clean markdown report with:
See references/model-pricing.md for current pricing across all providers.
Update this file when prices change.
See references/task-classification.md for detailed heuristics
on how tasks are classified into complexity tiers.
Generated Mar 1, 2026
A SaaS startup uses multiple AI agents for customer support, content generation, and system monitoring. They want to audit their setup to reduce overspending on high-cost models for simple tasks like status checks and notifications, reallocating budget to more complex development tasks.
An e-commerce platform employs AI agents for product description generation, inventory alerts, and customer sentiment analysis. They need to identify wasteful token usage in cron jobs and ensure model-task fit to improve ROI without compromising quality on critical tasks like security reviews.
A financial services firm uses AI agents for regulatory reporting, fraud detection, and market analysis. They require an audit to verify that complex tasks like compliance checks use appropriate high-tier models while downgrading simple health monitoring tasks to cut costs safely.
A healthcare organization leverages AI agents for patient data summarization, appointment reminders, and research analysis. They seek to optimize costs by classifying task complexities, ensuring sensitive tasks like security reviews retain top models while routine notifications use cheaper alternatives.
A media agency utilizes AI agents for drafting articles, social media posts, and performance analytics. They aim to audit their setup to balance model usage, downgrading repetitive content drafts to mid-tier models and reserving high-tier models for complex creative writing and coding tasks.
Companies offer AI-powered tools via monthly subscriptions, using agents for automated features. This audit helps them optimize internal agent costs, improving profit margins by reducing wasteful spending on model usage across customer-facing and backend tasks.
Consulting firms integrate AI agents into client systems for automation and analytics. They use this skill to audit client setups, providing cost-saving recommendations as a value-added service, enhancing client retention and upselling opportunities.
Platforms provide free basic AI agent tools with premium features for advanced usage. This audit assists in managing infrastructure costs by optimizing model allocations, ensuring free-tier users don't incur high expenses on simple tasks while premium features use appropriate models.
š¬ Integration Tip
Run the audit in dry-run mode first to preview analysis without changes, and schedule monthly audits via cron jobs to track cost trends and adjust recommendations as usage evolves.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Search and analyze your own session logs (older/parent conversations) using jq.
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linking related objects, enforcing constraints, planning multi-step actions as graph transformations, or when skills need to share state. Trigger on "remember", "what do I know about", "link X to Y", "show dependencies", entity CRUD, or cross-skill data access.
Ultimate AI agent memory system for Cursor, Claude, ChatGPT & Copilot. WAL protocol + vector search + git-notes + cloud backup. Never lose context again. Vibe-coding ready.
Headless browser automation CLI optimized for AI agents with accessibility tree snapshots and ref-based element selection