Find wasted AI spend in OpenClaw, Hermes, Claude Code, Cursor, and any framework that can export a JSON event payload.

Xerg is a local-first CLI for auditing AI spend in dollars, not raw token counts. It reads OpenClaw logs/transcripts or an independent sanitized trace capture, Hermes v0.17+ state.db, Claude Code transcripts, and Cursor usage exports — plus event payloads from any framework via xerg ingest — separates confirmed waste from savings opportunities, and lets you measure fixes with --compare.

Everything runs locally by default. The CLI is publicly installable from npm as @xerg/cli, but it is not open source. No account is required for local audits. A free hosted workspace keeps the last 30 days of pushed audits; hosted MCP requires a Pro or Enterprise workspace.

The npx @xerg/cli@latest path fetches and executes the published npm package before running Xerg. If you want to avoid that fetch on each use, install the CLI globally with npm install -g @xerg/cli.

Install this skill into any agent that supports the Agent Skills standard (Claude Code, Codex, Cursor, and others):

npx skills add xergai/skills

Then ask your agent to audit your AI spend. The agent runs the CLI itself and explains the findings.

Install the CLI directly:

npm install -g @xerg/cli

Or run without installing:

npx @xerg/cli@latest init

• Retry waste - failed calls that burned spend before a later success
• Loop waste - runs that exceeded efficient iteration bounds
• Context bloat - input token volume far above the workflow baseline
• Downgrade candidates - expensive models on operationally simple tasks
• Idle waste - recurring heartbeat or monitoring loops worth reviewing

Local JSON findings can include signalSource, ruleId, and evidence references so agents can distinguish observed signals from inferred or legacy unknown provenance. Compare output leads with normalized waste rate and per-unit rows before workload-dependent spend deltas.

Interactive first run (humans):

xerg init
xerg audit --compare

Non-interactive path (agents, scripts, CI):

xerg doctor
xerg audit --json
xerg audit --json --compare
xerg collect openclaw
xerg audit --otlp-file ./openclaw.capture.jsonl

Add --runtime openclaw, --runtime hermes, or --runtime claude-code when more than one runtime is detected.

• Local machine: OpenClaw, Hermes, and Claude Code (xerg audit --runtime claude-code)
• Independent local OpenClaw traces: xerg collect openclaw or xerg audit --otlp-file <capture.jsonl>
• Local Cursor usage export: xerg audit --cursor-usage-csv ./cursor-usage.csv
• Any framework's exported event payload: xerg ingest --file payload.json
• Remote OpenClaw sources via SSH or configured remote transports

If local defaults are empty, inspect the target directly first with xerg doctor --remote user@host.

xerg connect
xerg mcp-setup

• connect offers browser auth and pushing the latest audit
• mcp-setup prints or writes hosted MCP config for supported clients
• local audits and compare remain available if you skip hosted setup

• Local audits read OpenClaw, Hermes, Claude Code, Cursor usage, or ingest payload files and may write local JSON snapshots for --compare.
• Hermes uses ~/.hermes/state.db read-only by default. Optional observer telemetry and the complete mechanical-efficiency block remain local and are excluded from push payloads.
• OpenClaw trace collection binds only to loopback, sanitizes before persistence, receives traces only, never merges with transcripts, and never pushes automatically. toolActivity and workloadEconomics remain local.
• Remote OpenClaw audits pull selected files to local temporary storage before analysis.
• Xerg Cloud sync only happens when you run connect, audit --push, or push.
• Push payloads include audit totals, rollups, findings, recommendations, comparison deltas, and source metadata. They exclude raw prompt and response content, local source file paths, local snapshot store paths, and internal finding details.
• Rollup-level provenance (per-finding signalSource, waste-by-signal-source totals, and pricing coverage counts) is carried on the wire; evidence references and internal details stay local.

xerg audit --push --fail-above-waste-rate 0.25
xerg audit --fail-above-waste-usd 100
xerg audit --json

• Docs: xerg.ai/docs
• Service status: status.xerg.ai
• Skill: xerg.ai/skill.md
• npm: @xerg/cli
• Pricing: xerg.ai/pricing
• Support: query@xerg.ai

This skill is maintained by Xerg and synced to the public xergai/skills repo from the Xerg release train, versioned with @xerg/cli. The skill content is free to install and use with any agent; the Xerg CLI it drives is publicly installable from npm but not open source.