Daniel San
@dani_avila7
Followers
16K
Following
3K
Media
945
Statuses
2K
co-founder and CTO, building @aitmpl_com + @codegptAI + @deepgraphMCP | Powered by TypeScript & Pumpkin Spice Lattes ☕️
Grand Rapids, Michigan, USA
Joined January 2015
What exactly happened in the Vercel Sandbox during the review? Stats show "Files read: 9, Suggestions made: 2" but I'd love more visibility into: Which specific files were analyzed? What linters ran? Were any tests executed? Having access to these logs (or a detailed summary)
1
0
1
MODERATOR: Are you willing to commit to NOT raise the sales tax? MIKIE SHERRILL: I'm not going to commit to anything right now. On Nov. 4, vote NO on Mikie Sherrill. ❌
48
116
378
The agent found 2 shell injection vulnerabilities (nice! 👏), but suggested different fixes for the same issue. I think it should recommend only the best solution to avoid confusion Also, can you add severity levels ??(Critical/High/Medium/Low) and if issues block the merge or
2
0
2
Hey @JohnPhamous Here's my feedback after running a couple of executions. The check status stayed on "Thinking..." even after the review completed. Would be better to show a clear final status like "Review completed"
Testing @vercel Agent Review and it's working great 👏 Full codebase analysis, existing comments review, and prioritized findings. Saw @rauchg post last week and had to try it
2
0
5
Just automated Claude Code Templates releases to Discord. Working on integrating Claude Agent SDK so Claude can handle repo management and community communication - issues, PRs, discussions, all automated. LLMs can plug into these parts of your dev workflow and handle the
1
1
29
Adults with neurodiversity can face mobility challenges that limit their independence & employment. See how Waymo could help.
0
5
33
What really excites me is Vercel Sandbox... interesting possibilities for Claude Code workflows. Will experiment with this during the week
2
0
6
IT FINALLY ARRIVED 🇺🇸 5 years of paperwork, immigration appointments, endless waiting… it’s over. I’m officially a U.S. citizen. I can finally travel freely. Started from scratch, built 2 businesses, bought a home, raised a beautiful family. This is just the beginning of my
22
3
132
I get the feeling this is going to happen more and more often. What do you think?
Started using "Claude Code" today. I'm impressed ❤️ Not sure if I’m going back to Cursor 🤔
11
0
45
The npm for CC AI components 📦
めっちゃ良いサイト見つけた。 Claude Code用のAIエージェントが100以上掲載されている「Claude Code Templates」UXデザイナーやSQLエンジニアなどからプロンプトエンジニアや法務専門など文系職向けもあり有益。HooksやMCPも多数まとめられている👇🧵
0
0
10
A few weeks ago, Anthropic mentioned my opensource project 👏 I started this project on weekends just to explore Agentic programming and Claude Code. It’s surreal to see something I built purely to learn become part of such an incredible community. The repo is about to hit
10
2
82
Common question: Do https://t.co/pSU4AtqPi9 components work with Gemini CLI, Codex, AmpCode or other tools? The answer is no. Here's why 👇 Different models = different pre-training = different behavior. LLMs aren't REST APIs, you can't swap models and expect the same results.
@dani_avila7 Will it work in Gemini cli?
6
2
30
A few weeks ago, Anthropic mentioned my opensource project 👏 I started this project on weekends just to explore Agentic programming and Claude Code. It’s surreal to see something I built purely to learn become part of such an incredible community. The repo is about to hit
10
2
82
I haven’t run /init again lately. I’m worried a rewrite might lose important details. Now I just ask Claude Code to update CLAUDE. md with new workflows, or I update it directly myself. This markdown file is so critical to your project’s success that you should know it inside
@dani_avila7 I've know this but been unable to practice it. Do you ever update your https://t.co/WemPCPpNVA file after the initial /init?
3
0
15
Done! One single command and the whole setup is ready Just run: npx claude-code-templates@latest --agent deep-research-team/nia-oracle --mcp "deepresearch/mcp-server-nia" If you're building tools on top of Claude Code, make sure distribution is as seamless as possible.
16
36
345
Best way to debug Claude Code sessions and inspect agent context. Just Run: npx claude-code-templates@latest --chats Browser interface with advanced search across all your sessions (name, ID, directory, dates, message content). Shows every tool call with exact JSON executed.
12
40
351
Big news! Anthropic opening in Seoul. Korean users already rank top 5 globally in Claude usage. Seoul’s developer community is massive. 👏👏 Shoutout to the Korean community that follows me! 🇰🇷
We're opening our third Asia-Pacific office in Seoul, Korea. Korean users of Claude rank in the top 5 globally. We can’t wait to open an office in Gangnam and begin hiring a local team. Read more:
1
1
12