savings — Claude Code skill: savings reports and installation
Track savings from routing tasks to cheaper models, review routing quality, and prepare a weekly spend digest.
npx skills add ypollak2/llm-router --skill savings -a claude-codeWhat savings does
Use /savings to view savings across sessions for today, a week, a month, or all time. The available views also cover usage, Codex status, and providers; a web dashboard opens at localhost:7337.
A weekly digest can be printed or sent to Slack, Discord, or a generic JSON webhook. Sending requires LLM_ROUTER_WEBHOOK_URL in the environment; without it, you can still print the digest. The digest flags a daily spend spike when it exceeds twice the 7-day average.
The skill can show the active routing policy and the last 10 audit decisions, and report per-task routing accuracy based on 👍/👎 feedback. For full routing statistics, run llm-router routing-report in a terminal.
What it helps with
- check savings across a chosen time period
- spot daily spend above the 7-day average
- send a weekly savings digest to a team channel
- inspect active routing rules and recent audit decisions
- review routing accuracy using your feedback
Example prompts for the agent
- Show me how much routing saved this month.
- Prepare and send the weekly savings digest to the configured channel.
- Show the active routing policy and recent audit decisions.
How to install savings
1Skills CLI (Vercel)
npx skills add ypollak2/llm-router --skill savings -a claude-code2Manually: put the files in place
~/.claude/skills/savings/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/savings/
1Skills CLI (Vercel)
npx skills add ypollak2/llm-router --skill savings -a codex2Manually: put the files in place
.agents/skills/savings/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add ypollak2/llm-router --skill savings -a cursor2Manually: put the files in place
.cursor/skills/savings/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install ypollak2/llm-router savings2Manually: put the files in place
.github/skills/savings/SKILL.md1Skills CLI (Vercel)
npx skills add ypollak2/llm-router --skill savings -a gemini-cli2Platform CLI
gemini skills install https://github.com/ypollak2/llm-router --path skills/savingsManually: put the files in place
.devin/skills/savings/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add ypollak2/llm-router --skill savings -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about savings
How do I install savings in Claude Code?
Run npx skills add ypollak2/llm-router --skill savings -a claude-code. The skill file is at ~/.claude/skills/savings/SKILL.md.
What does savings do?
It reports savings from routing work to cheaper models, along with spend summaries and routing-quality information.
What is needed to send a weekly digest to Slack or Discord?
Set LLM_ROUTER_WEBHOOK_URL in your environment, then request a weekly digest with sending enabled. The skill identifies the webhook type from its URL.
Source text
SKILL.md · MIT
---
name: savings
description: Report how much routing has saved by sending tasks to cheaper models
trigger: /savings
---
# LLM Router — Savings Tracking Skill
Track and report how much you've saved by routing tasks to cheaper models.
## Viewing Savings
```
llm_savings — Cross-session savings by period (today / week / month / all time)
llm_usage("all") — Full dashboard: subscription %, Codex status, savings, providers
llm_dashboard — Open web dashboard at localhost:7337
```
## Savings Digest (Slack / Discord)
Send a weekly savings summary to your team channel:
```
llm_digest(period="week") — format digest, print only
llm_digest(period="week", send=True) — format + push to LLM_ROUTER_WEBHOOK_URL
```
Set `LLM_ROUTER_WEBHOOK_URL=https://hooks.slack.com/...` in your environment.
Auto-detects Slack, Discord, or generic JSON webhook from the URL.
## Spend Spike Alerts
The digest automatically flags when today's spend is > 2× the 7-day average.
No config needed — it's always on.
## Routing Policy
```
llm_router_admin(action="policy") — Show active org/repo routing policy + last 10 audit decisions
```
Set policy in `~/.llm-router/org-policy.yaml`:
```yaml
block_models:
- "o3*" # never use o3 series (too expensive)
- "gpt-4o"
allow_models:
- "gemini*" # prefer Gemini (allow overrides block)
task_caps:
image: 2.00 # max $2/day on image generation
```
## Routing Quality
```
llm_benchmark — Per-task routing accuracy from your 👍/👎 feedback
llm_rate — Rate the last response to improve future routing
```
Full routing stats / classifier accuracy: no MCP door on the consolidated surface —
run `llm-router routing-report` from a terminal.
© ypollak2 · MIT · ypollak2/llm-router
