SafeReceipt
Every agent action leaves a receipt.
safereceipt.vercel.app SourceAI Models & Agents / Prompts, Evals & LLM Observability
- Monthly visits
- <5K
- Oct 2026
- Public revenue
- Undisclosed
- No public revenue evidence yet
- Entry price
- Free
- No monthly subscription pricing
- Launched
- 9mo ago
- 2026-02-19 · Est.
Traffic
Marketing moves
Launches, search & content, referrals, promotionsLaunched on 阮一峰周刊投稿Launch
【开源自荐】SafeReceipt:给 AI agent 的每个动作开一张收据,谁都能自己核对 — - 仓库:https://github.com/calderbuild/SafeReceipt - 在线体验:https://safereceipt.vercel.app/fleet (不用装钱包) 这周我让 AI 帮忙重写这个项目的 README,它没打开合约源码,凭记忆写了 4 个函数签名,写得很自信,看上去完全没毛病。后来对着源码一查,4 个全错:有一个参数顺序反了,有一个凭空多了个参数。 这件事刚好就是我在做的东西想解决的:agent 说自己做了什么,和它实际做了什么,中间差多少,现在基本只能看运营方自己的日志,而日志想改就能改。 SafeReceipt 的做法是给 agent 的每个动作开一张收据: 1. 动手之前,把…
Sep 26 · 52d agoLaunch page
At the time: <5K visits/moLaunched on V2EX (4 comments)Launch
分享一个 AI Agent 问责协议: SafeReceipt,用链上收据证明 AI 到底做了什么 — 背景:AI Agent 帮你执行链上交易越来越常见了——授权 ERC20 、自动转账、和 DeFi 合约交互。 但有个问题:Agent 说它做了 X ,但如果实际做的是 Y ,你怎么证明?事后看 tx 记录?那你怎么证明你当时的意图不是 Y ? 做了一个叫 SafeRecei…
Feb 20 · 9mo agoLaunch page
Launched on 阮一峰周刊投稿Launch
【开源自荐】SafeReceipt -- AI Agent 问责协议,链上收据证明 AI 到底做了什么 — ## SafeReceipt — Agent Accountability Protocol 当 AI Agent 替你操作链上交易时,你怎么知道它有没有偷偷改金额、换收款地址? SafeReceipt 解决这个问题:**在 AI 执行交易之前,先把用户的真实意图做成密码学收据上链;执行之后,对比收据和实际交易,自动判定是否一致。** 一句话:**When AI fails, receipts prove who's responsible.** **在线体验:** https://safereceipt.vercel.app **源码仓库:**
Feb 19 · 9mo agoLaunch page
Open Source & Agent Ecosystem
- GitHub Stars
- Public repo
- Forks
- —
- License
- See repository
- Primary language
- —
Agent, MCP & Developer Integration
CLI / SDK- Exposed tool capabilities:
- Agent Audit TrailsAgent Outcome VerificationERC20 approveBeginAction()Emit()EndAction()
Traffic sources
Traffic breakdown
Channels
- Direct43%
- Organic Search31%
- Referrals12%
- Social6%
- Other7%
Countries
- Taiwan100%
Search
Domain authority —Domain registered 2020-01-28
No search keyword data for this website yet.
Traffic data: Vaneform
Pricing
Pricing page not identified yet, or no public pricing listed.
Pages & changes
Key pages
Weekly snapshotsChanges
No pricing, plan, or positioning changes detected since tracking started.
Questions & evidence
- How much website traffic does SafeReceipt get?
- Vaneform estimates <5K website visits in 2026-10. Visits can include repeat visits; this estimate measures web traffic separately from paying customers or revenue.Source
- What is SafeReceipt and what platforms does it support?
- SafeReceipt is catalogued under Prompts, Evals & LLM Observability: Every agent action leaves a receipt.. Supported platforms: CLI / SDK.
- Is SafeReceipt open source?
- SafeReceipt has a public repository (https://github.com/calderbuild/SafeReceipt) with its source repository published.Source
Related products in this category
These products share a catalog category. Compare customer tasks, features and deployment requirements before treating them as alternatives.
| Product | Entry price | Visits | Revenue |
|---|---|---|---|
| LMSYS Chatbot ArenaGiantExperience the frontier | — | 35.8M | Undisclosed |
| Prompt CowboyMatureTurn lazy prompts into great ones | $8/mo | 1.1M | Undisclosed |
| LiveBench 2.0MatureA challenging, contamination-free long-horizon AI benchmark. | — | 688.3K | Undisclosed |
| Codex雷达站MatureObserves Codex model tiers, speed, and real subscriber capability changes. | — | 605.1K | Undisclosed |
| LangfuseMatureOpen Source Agent Evals &Observability | $29/mo | 586K | Undisclosed |
| SmithMatureTake agents from prototype to production. LangSmith gives you the tools to build, debug, evaluate, and ship reliable agents. | — | 566.4K | Undisclosed |