Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/jansenanalytics/claudex/cost-optimizernpx skills add JansenAnalytics/claudex --skill cost-optimizergit clone --depth 1 https://github.com/JansenAnalytics/claudexWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00029 | $0.00384 |
| Opus 5 | $0.00015 | $0.00192 |
| Sonnet 5 | $0.00006 | $0.00077 |
| Haiku 4.5 | $0.00003 | $0.00038 |
Grade A, and why
cost-optimizer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
cost-optimizer
Track and optimize LLM token spend: estimate costs per task, recommend model tiers, monitor cumulative usage, suggest batching strategies.
Use When
- Analyzing API costs for a project or session
- Choosing between model tiers (opus vs sonnet vs haiku)
- Estimating token budgets for upcoming work
- Reducing unnecessary token usage
- Reviewing cumulative spend
Scripts
token-counter.py
Count tokens in text input (stdin or file). Uses tiktoken-compatible estimation.
echo "Hello world" | python3 scripts/token-counter.py
python3 scripts/token-counter.py < myfile.txt
python3 scripts/token-counter.py --file myfile.txt
cost-estimator.py
Map token counts to USD costs for various models.
python3 scripts/cost-estimator.py --input 1000 --output 500 --model claude-opus-4
python3 scripts/cost-estimator.py --input 1000 --output 500 # shows all models
usage-analyzer.sh
Analyze OpenClaw usage logs to find spending patterns.
bash scripts/usage-analyzer.sh [--days 7] [--log-dir ~/.openclaw/logs]
References
references/model-pricing.md— Current pricing for major LLM providersreferences/token-heuristics.md— Rules of thumb for estimating tokensreferences/cost-reduction.md— Strategies to reduce token spend
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 55 lines · 29 tokens per session scan A 08ddadac088f
cost-optimizer is a skill published in the GitHub repository JansenAnalytics/claudex (5 stars, last pushed 2mo ago), licensed MIT. It adds 29 tokens to every session and 384 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
torchforge-rl-training
Provides guidance for PyTorch-native agentic RL using torchforge, Meta's library separating infra from algorithms. Use when you want clean RL abstractions, easy algorithm experimentation, or scalable training with Monarch and TorchTitan.
llamaguard
Meta's 7-8B specialized moderation model for LLM input/output filtering. 6 safety categories - violence/hate, sexual content, weapons, substances, self-harm, criminal planning. 94-95% accuracy. Deploy with vLLM, HuggingFace, Sagemaker. Integrates with NeMo Guardrails.
prompt-guard
Meta's 86M prompt injection and jailbreak detector. Filters malicious prompts and third-party data for LLM apps. 99%+ TPR, <1% FPR. Fast (<2ms GPU). Multilingual (8 languages). Deploy with HuggingFace or batch processing for RAG security.
prompt-engineering
提示工程技术与 Agent 设计模式.
Prompt安全防护
为大语言模型应用部署 Meta 的 86M 参数提示注入/越狱检测器。真阳率 99%+，GPU 推理延迟 <2ms，支持多语言 RAG 安全防护。.
prompt-modeler
Generate structured strategic prompts with diagnostic and multiple options. Triggers on: /modelar-prompt.