Pinned Tweet
I was constantly fighting Claude Code's context limits. So many /clear and /compacts!
So I built a system of subagents. Now my sessions run 10x longer, and the code quality is higher.
Here are 6 battle-tested subagents that transformed my workflow (with working code) 🧵👇
Incredible news from Anthropic today. Browser use AND Claude Code on the web! 🚀
Research preview of Claude Code on Claude[.]ai
- updated layout with the prompt composer and sessions moved to the sidebar
- repository setup requires installing the GitHub Claude app on your repository and committing the "Claude Dispatch" GitHub workflow file
- you can enable email and web notifications for Claude Code updates
1
902
A new class of malware targeting LLM-assisted coders appears! 🤯
Will a new class of security tools will have to deal with this?
A popular NPM package got compromised, attackers updated it to run a post-install script that steals secrets
But the script is a *prompt* run by the user's installation of Claude Code. This avoids it being detected by tools that analyze code for malware
You just got vibepwned
1
4
948
Straight from Anthropic.
For sure I agree compacting often causes messy context. I think /clear could be better with handoff though so the agent isn’t starting from zero.
🫸 STOP COMPACTING, use /clear
the biggest difference I see between how the Claude Code team uses CC and the broader community how often they run into compacts
instead of keeping long running chats open across multiple tasks, when you finish a task use /clear & start again
2
5
1,159
Most Claude Code hooks are written in bash. That's a mistake.
I rebuilt 4 essential Claude Code hooks in Python instead - no dependencies, cross-platform, actually readable.
Here's why this matters and the 4 hooks that'll power-up your Claude Code workflow... 🧵👇
I've put all 4 Python hooks, with full source code and setup instructions, in a single Gist linked below
Grab the complete 'Python Power Pack' here:
gist.github.com/jeremynsl/25…
1
1
100
⚡️ Next Up: I'm building a 'Context Keeper' hook to solve one of Claude's biggest limitations (losing context on /clear). It's coming soon™️.
Reply with your #1 Claude Code frustration and I'll DM the finished 'Context Keeper' to everyone who shares a pain point. 👇
1
1
99
Long conversations in @googleaistudio have been a bit of a nightmare to scrollback through... 😫
Looks like an update just pushed out - now each convo 'turn' is a node in the scrollbar, which shows a summary of what was discussed! What do you think - did they cook?
2
110
LLM Benchmarks are such a mess... So Mistral Medium 3.1 is apparently #1 on @lmarena_ai . But on an aggregated suite of other benchmarks it ties Gemini 2.0 Flash for dead last 🤔 And cost of running benches was 8x more 💰than 2.0 Flash.
Anyone tested this model themselves yet?
Mistral Medium 3.1 just landed on @lmarena_ai leaderboard—punching way above its weight!
🏆 #1 in English (no Style Control)
🏆 2nd overall (no Style Control)
🏆 Top 3 in Coding & Long Queries
🏆 8th overall
Small model. Big impact. Try it now on Le Chat and the API!
2
19
3,842
Claude Code 1.0.89 dropped on auto-update. No changelog yet. But I see token count in my statusbar!
3
12
818
Jeremy Janzen retweeted
Oldie but only found out last night. Claude's search is using the Brave API. This was confirmed at the Vercel closing Keynote. techcrunch.com/2025/03/21/an…
#PromptEngineering vs #PromptSynthesizing ❓
Working on something new. Demo next week.
1
5
376


















