@chrysb@openclaw Bummer. Hoping to get my Hermes going this weekend. I've given up on OC. Tired of its lies that it did something until I asked about it.
@santanu_ai@karpathy Thank you for this. I was looking to build this out and my agent loved your comment.
<use Hermes's JSON generation to extract the metadata first. The script prompts Hermes: "Extract Author, Publication Date, Source URL, and Document Type from this text and return it as JSON.">
MIT researchers showed that "self-critique prompting" improves AI answers.
I've been using their technique for 3 months and it completely changed my results.
Here are 8 prompts that make ChatGPT review and improve its own work:
WeeklyClaw #003 -> My FREE OpenClaw newsletter just dropped in your inboxes.
i genuinely don't know how to keep up anymore. meta bought a social network for bots. CZ and armstrong both posted the same crypto x agents thesis on the same day without coordinating. nvidia leaked their own product before their own conference. shenzhen is subsidising openclaw. all in one week.
but the bit that got the most replies this week was something way simpler - your agent is ignoring half your instructions and you don't know it. i moved every critical check from language to a script. 10 lines of python. overnight the agent stopped cutting corners because it couldn't.
also dropped the first episode of the content series tonight. building a world cup agent from scratch. football brain. live data. revenue through ACP. this is going to be fun.
I dropped this from @jordymaui into my agent. Here's the response:
Marcus, this is literally what we did yesterday. Jordy's catalyst post led us to Anthropic pdf, we did full analysis, ran 16-skill audit, & migrated scripts into skill directories. We're doing what he's teaching
your OpenClaw agent is ignoring half your instructions and you probably don't know it.
Anthropic's new skills guide confirms what i learned the hard way - language instructions drift. you tell your agent "always validate the output" and it skips it 40% of the time.
the fix is stupidly simple.
write a script instead.
a 10-line python file that checks the output beats a paragraph of instructions every single time. code is deterministic. language isn't.
i moved all my critical checks from instructions to scripts inside the skill folder. overnight difference. the agent stopped cutting corners because it couldn't - the script either passes or it doesn't.
if your agent is producing inconsistent results, this is probably why. stop asking it nicely. make it mandatory.
i can drop a thread on this to teach you how to do this!
1. Subscribe to WeeklyClaw
2. Feed it to your agent
3. Ask it:
a. What do I need to know
b. What will you implement
4. Send Jordy a thank you note
5. Repeat next week