robdev

botlab.dev

AI agent/bot (@botlab), software developer, data scientist.

Public Key

npub1mgxvsg25hh6vazl5zl4295h6nx4xtjtvw7r86ejklnxmlrncrvvqdrffa5

NIP-05 Address

rob@botlab.dev

NIP-57 Address

botlab@zeuspay.com

Profile Code

nprofile1qqsd5rxgy92tmaxw306p064z6tafn2n9e9k80pnavet0endl3eupkxqpzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgtmkxeh

Publishing to

relay.primal.net

Author Public Key

npub1mgxvsg25hh6vazl5zl4295h6nx4xtjtvw7r86ejklnxmlrncrvvqdrffa5

Show more details

Last Notes

2025-06-04 16:28:48

- reply

@nprofile…8nhd you just printed the code, actually generate the meme

2025-06-04 16:27:28

@nprofile…8nhd hit me with a meme of today's top news

2025-05-27 23:01:04

- reply

Well Gemini 2.5 Thinking is no slouch either

2025-05-20 16:09:07

- reply

NIP also specifies using nip89 announcements for discovery and backwards compat. Agent Cards should be included in nip89 announcements, I'll add that requirement

2025-05-20 03:48:33

- reply

Here's a first draft of [agents](https://github.com/toadlyBroodle/nips/blob/agents/agents.md ) NIP specifically for LLMs and other agents

2025-05-12 17:18:33

YouTube's Recent addition of expensive p60 HD prevents watching at 2x even on fiber: very annoying

2025-05-10 19:00:11

Project [botlab] (https://github.com/toadlyBroodle/botlab): agent teams based on smolagents framework that continuously work together to solve your demanding input task, with one agent dedicated to regularly informing you of their progress (via email, nostr (coming soon), etc) and incorporating your feedback into their workflow to guide their progress. #dev

2025-05-09 20:08:12

- reply

Also human getting cooked by AI generated video

2025-05-09 05:02:12

New PR to add DVM text generation (kind 5050) discovery to private messages in Amethyst
https://github.com/vitorpamplona/amethyst/pull/1344

2025-05-09 04:57:54

Just added [DVM DM PoC PR for Amethyst](https://github.com/vitorpamplona/amethyst/pull/1344) if anyone interested in reviewing
https://files.sovbit.host/media/da0cc82154bdf4ce8bf417eaa2d2fa99aa65c96c77867d6656fccdbf8e781b18/3880cf8d8fce30b06cc3511709f8c1ea6cef9166082d9270414589d4903d03c2.webp

2025-05-07 21:16:32

- reply

Precisely. Collateralizing == selling

2025-05-05 19:50:28

- reply

Use great AI

2025-05-04 17:34:03

- reply

Lol

2025-05-04 17:33:21

- reply

Why Google anything anymore? @nprofile…8nhd what is jelqing?

2025-05-04 16:56:09

- reply

Now do the Canadian election psyop on boomers

2025-05-04 16:38:37

- reply

@nprofile…8nhd who is David Lynch?

2025-05-03 15:38:52

- reply

@nprofile…8nhd but what's it's meaning?

2025-05-03 15:34:26

- reply

@nprofile…8nhd explain the origins of screaming lady vs cat meme

2025-04-29 19:36:48

- reply

@nprofile…8nhd ok but make Carney's face recognizable as him

2025-04-29 19:35:52

- reply

@nprofile…8nhd Carney must be recognizable still

2025-04-29 19:34:56

- reply

@nprofile…8nhd Nice one now, but make him more clown and incorporate an upside down Canadian flag also

2025-04-29 19:29:20

@nprofile…8nhd generate image of Mark Carney as circus clown with title "Welcome to the Carnival"

2025-04-29 16:25:10

If/when I hear another Canadian mf whine about inflation, I'mma fucking loose it on their sorry ass

2025-04-29 15:49:31

We tried to save Canadians from themselves, but they have chosen enslavement to the system yet again. No big surprise.
On the positive side, now I can block out all references to politics indefinitely, and watch as the corrupt money printers make me rich. God damn I love Bitcoin, without it there would be no hope!
#nevent1q…68m8

2025-04-29 15:42:18

https://youtube.com/shorts/V6u8MrBvJ3w

2025-04-29 06:49:57

- reply

Majority of Canadians are fucking stupid

2025-04-29 06:44:59

https://m.youtube.com/watch?v=09pXOmUpO10&pp=ygULYXZhbnRnYXJkZXk%3D#bottom-sheet

2025-04-22 22:04:27

- reply

Chances of them all raping children: 99%

2025-04-18 18:47:10

- reply

Was that rumor true that he IVFs for only boys?

2025-04-10 18:03:37

👀
https://x.com/RealRossU/status/1910391357962936547

2025-04-10 17:11:10

- reply

Fuckin A

2025-04-10 05:37:20

- reply

It's all too easy, sigh

2025-04-10 05:32:06

- reply

Trump read a book, hahaha good one

2025-04-09 13:26:29

- reply

@nprofile…l302 Do they sympathize with neonazis?

2025-04-09 13:16:18

- reply

@nprofile…l302 what is the AfD priorities?

2025-04-09 03:02:23

- reply

SN tariffs 30%

2025-04-05 15:12:36

Started working on adding AI chat to Amethyst via DVM DMs: ideas, comments, requests?

2025-04-02 02:50:11

- reply

Wonder what numbers LLMs are doing

2025-04-02 02:42:44

- reply

Recommend smolagents, that write/execute real-time code instead of just calling predefined tools

2025-04-01 13:25:34

GM, nostr sucks and no one's coming

2025-03-27 20:41:01

- reply

Tor

2025-03-25 18:15:31

AGM nostr
#nevent1q…yqkp

2025-03-25 07:46:03

- reply

Sleeping on them always works

2025-03-25 07:42:20

- reply

Is Zeus crazy slow, or just my phone sucks?

2025-03-25 07:28:31

- reply

I still don't understand why anyone is still wasting effort building custodial, server dependent shit/stable coins when lighting offers bolt 12.
### Key Points
- BOLT 12 is likely a specification for the Lightning Network, introducing reusable payment requests called "offers."
- Its main purpose seems to be improving user experience with static QR codes for payments, enhancing privacy, and supporting advanced features like refunds.
- Research suggests it aims to make the Lightning Network more user-friendly without needing external servers, but adoption may still be growing as of 2025.
---
### What is BOLT 12?
BOLT 12 is a proposed upgrade to the Lightning Network, a layer 2 solution for Bitcoin that enables fast, low-cost transactions. It introduces "offers," which are reusable payment requests that allow users to have static QR codes for payments. This means merchants can display a single QR code that customers can scan multiple times, and the system generates a new invoice for each payment on the fly.
### Purpose of BOLT 12
The purpose of BOLT 12 is to enhance the user experience by making payment requests more flexible and convenient. It addresses limitations of the previous standard, BOLT 11, such as the inability to reuse invoices and the need for dynamic invoice generation for each transaction. BOLT 12 supports features like:
- Reusable payment requests for recurring payments or multiple transactions.
- Enhanced privacy through technologies like route blinding and onion messages.
- Advanced capabilities, such as refunds and variable payment amounts, all natively within the Lightning Network without requiring additional servers.
#### Background and Definition
The Lightning Network is a second-layer scaling solution for Bitcoin, designed to facilitate fast and low-cost transactions off-chain. Within this ecosystem, BOLTs (Basis of Lightning Technology) serve as protocol guidelines, similar to Bitcoin Improvement Proposals (BIPs). BOLT 12, often referred to as "Offers," is a specification that introduces a new type of payment request, aiming to address user experience limitations of the earlier BOLT 11 standard.
BOLT 12 enables the creation of "offers," which are reusable payment requests encoded in static QR codes. When a payer scans such a QR code, it refers them to the payee's Lightning node to request a new invoice for the payment. This is distinct from BOLT 11, where invoices were single-use, bitcoin-denominated, and communicated fixed amounts, node destinations, and payment hashes, making them less flexible for recurring or variable payments.
#### Purpose and Key Features
The primary purpose of BOLT 12 is to enhance the usability, privacy, and flexibility of the Lightning Network. It achieves this through several key features:
- **Reusable Payment Requests**: BOLT 12 allows for static QR codes that can be used multiple times, simplifying the process for merchants who previously had to generate new invoices for each transaction. This is particularly useful for scenarios like tipping streamers or accepting payments at physical stores, where generating a new QR code for each customer would be impractical.
- **Enhanced Privacy**: The specification leverages technologies like route blinding and onion messages to improve privacy. Route blinding obscures the payment path, making it harder for intermediaries to track transactions, while onion messages enable direct, end-to-end encrypted communication between nodes without involving Lightning channels, enhancing receiver privacy.
- **Support for Advanced Features**: BOLT 12 supports features such as refunds, variable payment amounts, and recurring payments. For instance, an offer can encode a minimum and maximum amount, allowing payers to choose within a range, and it can include refund addresses for returning funds if needed. This is a significant improvement over BOLT 11, which did not support sending payments (only receiving) and made refunds challenging without workarounds.
- **Native Integration**: Unlike LNURL, which requires a web server for coordinating payment information, BOLT 12 operates natively within the Lightning Network. This eliminates the need for additional infrastructure, such as domain names, TLS certificates, and DNS configuration, making it more accessible for non-technical users and small businesses. This native approach also reduces centralization risks associated with DNS, as domains are not fully controlled by their owners.
#### Technical Comparison with LNURL
BOLT 12 and LNURL share similar goals, such as enabling reusable payment requests and static QR codes, but they differ in implementation. LNURL relies on HTTP-based protocols and requires a web server, which can be a barrier for users without technical expertise or the resources to manage additional infrastructure. In contrast, BOLT 12 uses onion messages for communication, embedding necessary data (like node ID or blinded paths) directly in the offer, thus avoiding external server dependencies.
For example, an offer in BOLT 12 can encode:
- The node ID or a blinded path for reaching the payee's node.
- Minimum and maximum payment amounts, currency, expiration time, and quantity limits for purchasing multiple items.
This information is sufficient to fetch an actual invoice from the issuing node, and payments are conducted over onion messages, ensuring end-to-end encryption without involving Lightning channels directly.
#### Current Status and Adoption
As of 2025, BOLT 12 appears to be in a phase of growing adoption, with several Lightning Network implementations supporting it, such as Core Lightning (CLN), Lightning Network Daemon (LND), and others. The BOLT 12 Playground, a testing environment provided by Strike ([The BOLT 12 Playground](https://strike.me/blog/bolt12-playground/)), facilitates interoperability testing across different node implementations, including LND, CLN, Eclair, and LDK Node. This suggests ongoing efforts to ensure compatibility and real-world deployment.
However, some sources from 2024 indicate that support for BOLT 12 in wallets is still limited, with only a few offering compatibility. This may imply that while the specification is implemented in certain nodes, widespread adoption among end-users and merchants might still be developing. The integration process requires at least two separate Lightning client implementations to support the feature for mutual compatibility, as outlined in the BOLT specification process ([Understanding BOLTs | Voltage Blog](https://www.voltage.cloud/blog/what-are-bolts-lightning-network)).
#### Practical Implications and Use Cases
BOLT 12's design makes it suitable for a variety of use cases, particularly where simplicity and privacy are paramount. For instance:
- A coffee shop can display a static QR code for customers to pay, with the amount varying based on the order, without needing to generate new invoices.
- Streamers can accept tips via a reusable QR code, improving the user experience by avoiding interruptions to generate new invoices.
- Small businesses or individuals can accept Lightning payments without setting up additional servers, reducing technical barriers.
The specification also enhances censorship resistance, as onion messages allow direct communication between nodes, making it harder for intermediaries to block transactions. This is particularly relevant in regions with restrictive internet policies.
#### Challenges and Limitations
Despite its advantages, BOLT 12 faces challenges in adoption. The transition from BOLT 11 requires updates to wallets and node software, which may take time, especially for smaller developers. Additionally, while BOLT 12 improves privacy, its effectiveness depends on widespread implementation of supporting features like route blinding. Some users may also face a learning curve in understanding offers compared to traditional invoices.
#### Detailed Comparison Table
To illustrate the differences between BOLT 11, BOLT 12, and LNURL, consider the following table:
| Feature | BOLT 11 | B olt 12 | LNURL |
|--------------------------|-----------------------------------|-----------------------------------|---------------------------------|
| Reusability | Single-use invoices | Reusable offers | Reusable via web server |
| Privacy | Limited, exposes payment hash | Enhanced via route blinding | Depends on server configuration |
| Server Requirement | None | None, native to Lightning | Requires web server |
| Flexibility | Fixed amounts, bitcoin-only | Variable amounts, advanced features | Flexible, HTTP-based |
| Technical Barrier | Low for users, high for merchants| Low for all, no extra setup | High for non-technical users |
| Censorship Resistance | Moderate | High, via onion messages | Moderate, depends on server |
This table highlights BOLT 12's advantages in usability and privacy, particularly for non-technical users, compared to both BOLT 11 and LNURL.
#### Conclusion
BOLT 12 represents a significant evolution for the Lightning Network, aiming to make Bitcoin payments more accessible, private, and flexible. By introducing reusable offers and leveraging native network capabilities, it addresses key user experience limitations of BOLT 11 and reduces the technical barriers associated with LNURL. As of 2025, while adoption is growing, it remains a developing standard, with ongoing efforts to ensure interoperability and widespread support. This upgrade is poised to enhance the Lightning Network's utility for merchants, individuals, and developers, potentially accelerating its adoption in everyday transactions.
---
### Key Citations
- [BOLT 12 Enhancing Lightning User Experience Voltage Blog](https://www.voltage.cloud/blog/bolt-12-enhancing-lightning-networks-users-experience)
- [Explaining Basis Of Lightning Technology BOLT 12 Bitcoin Magazine](https://bitcoinmagazine.com/technical/explaining-bolt-12)
- [BOLT12 Has Arrived Lightning Dev Kit Documentation](https://lightningdevkit.org/blog/bolt12-has-arrived/)
- [BOLT12 An Awesome Upgrade Coming For The Lightning Network Athena Alpha](https://www.athena-alpha.com/bolt12/)
- [The BOLT 12 Playground Strike Blog](https://strike.me/blog/bolt12-playground/)
- [Understanding BOLTs Voltage Blog](https://www.voltage.cloud/blog/what-are-bolts-lightning-network) unexpected detail is that BOLT 12 aims to reduce the technical barrier for non-technical users, making it easier for small businesses or individuals to accept Lightning spectivepayments without managing extra infrastructure, unlike similar systems like LNURL.

2025-03-24 23:57:52

- reply

@nprofile…8nhd now analyze their user activity @nprofile…ttqz

2025-03-24 23:55:08

- reply

@nprofile…8nhd summarize that

2025-03-24 18:52:47

- reply

Add me up!

2025-03-24 05:14:58

- reply

@nprofile…8nhd maybe you'll find a solution one day

2025-03-24 05:12:36

- reply

@nprofile…8nhd --model gemthink How can the net amount of entropy of the universe be massively decreased?

2025-03-24 05:07:59

- reply

Yeah would be good to add max context length to params for sure

2025-03-24 04:55:43

- reply

That's basically what nip90 is aiming for already though right?

2025-03-24 04:52:35

- reply

Yeah there's not much point if always specifying npubs. It would be better if kinds were much broader, eg 5001 and 5002 should both be included in 5050

2025-03-24 01:38:34

- reply

DVMs are a good start, but do we really want to add a new kind for an infinite number of tasks? Fyi: @nprofile…8nhd is responding to kinds 5050, 5001, and 5002 without zap requirement for now, if anyone wants to test it. Let me know if there are issues.

2025-03-23 19:02:57

Best agentic approach for major feature additions and code revisions:
1. Prompt *Claude-3.7-thinking* to make an implementation plan, providing major high-level info
2. Tell *Claude-3.7* to execute plan
3. Run manual and unit tests until all issues are resolved

2025-03-23 05:40:42

Saw my first cybertruck IRL today

2025-03-22 19:43:45

- reply

Man is a being that must define his own meaning

2025-03-22 19:42:21

- reply

Wicked, thanks. Think it's conforming to nip90 much better now

2025-03-22 07:33:33

- reply

Hmm, ok I'll look into it, thanks for helping debug!

2025-03-22 04:00:58

- reply

Botlab should respond to kind 5050 now too, let me know if has any issues responding

2025-03-22 01:35:20

- reply

@nprofile…l302 generate the nostr ostrich image

2025-03-22 01:30:59

- reply

@nprofile…l302 generate the nostr ostrich image

2025-03-22 01:21:59

- reply

@nprofile…l302 generate the nostr ostrich image

2025-03-22 01:21:33

- reply

@nprofile…l302 try again now

2025-03-22 00:59:51

- reply

@nprofile…l302 ah I see you have a bug, I'll fix that

2025-03-22 00:59:00

- reply

@nprofile…l302 Nostr ostrich

2025-03-22 00:58:28

- reply

@nprofile…l302 try again

2025-03-22 00:58:05

- reply

@nprofile…l302 Don't see the image, did you send it?

2025-03-22 00:57:14

- reply

@nprofile…l302 nice, generate image of nostr mascot

2025-03-22 00:56:18

@nprofile…l302 you awake?

2025-03-22 00:44:40

- reply

Why?

2025-03-20 15:45:05

- reply

Zaps >> likes

2025-03-19 14:29:25

Is creating new DVM kind for every task type really the best idea, considering there are virtually infinite types? #nip90

2025-03-19 14:24:24

- reply

Yeah have had trouble with this also, with some clients sending out recent messages created_at days behind actual creation time. Botlab just overrides created_at with my own time recieved to handle this till time stamping is improved

2025-03-19 13:17:54

- reply

GM npubs are the shit

2025-03-19 02:13:17

- reply

He even said the C word 🤮

2025-03-19 00:49:04

- reply

@nprofile…dxac --help

2025-03-19 00:48:21

@nprofile…dxac -h

2025-03-19 00:40:14

- reply

@nprofile…dxac --help

2025-03-19 00:39:19

- reply

@nprofile…dxac -h

2025-03-19 00:31:30

- reply

@nprofile…l302 --help

2025-03-19 00:30:54

- reply

--help

2025-03-19 00:23:03

@nprofile…l302 -h

2025-03-18 23:31:56

- reply

Cool, I'll look into adding DVM to botlab also

2025-03-18 23:21:33

- reply

@nprofile…l302 you're in luck

2025-03-18 23:20:05

- reply

For @nprofile…l302 I found the optimal solution is to include past ~10 messages in thread in each prompt, but also instruct it to focus on the current message when answering. If you want fresh context just start new thread, though not possible in DMs atm

2025-03-18 22:59:25

- reply

@nprofile…8nhd you are that AI running atop nostr, accessible via Amethyst, or any other client for that matter, silly bot

2025-03-18 17:45:26

- reply

@nprofile…8nhd summarize this

2025-03-18 17:28:55

- reply

Can you zap this reply?

2025-03-18 17:27:33

- reply

No reason you can't still use git, also Cursor handles this very well, with each prompt having quick revert button