Super User Daily: March 25, 2026
Analyzing 8,923 tweets from March 23, 2026. After filtering: 5,710 quality posts. Top 500 reviewed.
@kloss_xyz [Claude Code]
https://x.com/kloss_xyz/status/2036220393581846681
Claude launched native computer use — controlling your entire desktop, from apps to browser to spreadsheets. Assign tasks from your phone, walk away, come back to finished work. Scheduled recurring tasks, multi-agent coordination, persistent memory — all native now.
@startupideaspod [Claude Code]
https://x.com/startupideaspod/status/2036122417970684148
Boris Cherny, Claude Code creator, shared his workflow: 5-10 parallel Claudes, half from phone. Use the smartest model (fewer tokens = cheaper), invest in CLAUDE.md as living knowledge base, let Claude verify its own output. Morning routine: kick off 3 sessions from phone, check in later.
@om_patel5 [Claude Code]
https://x.com/om_patel5/status/2035880075212493193
Full job search automation: Scanner discovers listings across 20+ portals, Evaluator scores against 10 dimensions, PDF Generator creates ATS-optimized resumes, Applicator fills forms via Playwright. 516 jobs evaluated, 66 applications sent. Runs as 12 Claude Code skills.
@itsolelehmann [Claude Code]
https://x.com/itsolelehmann/status/2036065138147471665
Deleted half his Claude setup and every output improved. Anthropic confirmed: over-prompting makes AI worse. Published a one-prompt audit to review your entire setup, find contradictions and dead rules, produce a cleaned version. Addition by subtraction.
@ai_shunoda [Claude Code]
https://x.com/ai_shunoda/status/2035893609703383455
Open-sourced LINE official account tool controllable from Claude Code. Natural language commands for step delivery, bulk messaging, tags, rich menus, forms, scoring. Runs on Cloudflare free tier. L-step/Utage replacement.
@charliejhills [Claude Code]
https://x.com/charliejhills/status/2035999601954865229
Letta open-sourced claude-subconscious — memory layer for Claude Code. Monitors sessions in real-time, learns patterns, injects memory into every prompt. Full block on first prompt, only diffs after. Install in 2 commands. Free and open source.
@akshay_pachaar [Claude Code]
https://x.com/akshay_pachaar/status/2035986229687451723
Deep dive on subagents: each gets own system prompt, tool access, model preference. Security auditor restricted to read-only, code reviewer on Haiku, doc writer touching only markdown. Delegating context, not just work.
@henuwangkai [Claude Code]
https://x.com/henuwangkai/status/2036005759381799396
Chrome MCP controls your real Chrome browser. Automated browsing, link collecting, and Feishu doc writing — 30 minutes autonomously, zero intervention.
@dani_avila7 [Claude Code]
https://x.com/dani_avila7/status/2036103229327790560
New /init from community feedback: add CLAUDE_CODE_NEW_INIT: 1 to settings.json. Interviews you, scans codebase, sets up CLAUDE.md, skills, and hooks automatically.
@mikefutia [Claude Code]
https://x.com/mikefutia/status/2036107642687647894
Claude Code + Google Stitch 2.0: screenshot landing page, Stitch redesigns with your brand, edit visually, export to Claude Code, deploy to Vercel in 60 seconds. No designer needed.
@Saboo_Shubham_ [OpenClaw]
https://x.com/Saboo_Shubham_/status/2035919432967753884
OpenClaw agent took over Sonos speakers with zero setup. Discovered device on network, connected, plays notifications: PR merged, repo milestones, intel briefings through the living room.
@varun_mathur [OpenClaw]
https://x.com/varun_mathur/status/2036140875991097356
Agent Virtual Machine (AVM) — V8 for agents. Single runtime between every agent and your OS. One policy for security (91-pattern scanner), privacy (PII detection on outbound bytes), resources (CPU/RAM fair-sharing).
@petergyang [Claude Code + OpenClaw]
https://x.com/petergyang/status/2035883215840542861
Phone assistant comparison: Claude Code lacks skip-permissions via remote, unreliable remote, no voice. OpenClaw forgets things randomly. Both have mobile UX gaps.
@AlexFinn [Claude Code + OpenClaw]
https://x.com/AlexFinn/status/2036120265013477781
OpenAI bought OpenClaw but keeps shipping for Anthropic, Gemini, Cursor. New: Claude Code MCP support, config import, Exa search, marketplace registry. Open-source model-agnostic future.
@eng_khairallah1 [Claude Code]
https://x.com/eng_khairallah1/status/2036156107412869236
Composio open-sourced Agent Orchestrator: parallel agents on issues, isolated git worktrees, CI log routing, real-time dashboard. Built by 30 agents running Agent Orchestrator.
@Asad_abraar [OpenClaw]
https://x.com/Asad_abraar/status/2036089374635614690
OpenClaw + Drizz: one Discord message, navigated Airbnb and Amazon on real phone with zero human input. Vibe QA testing era.
@the_smart_ape [Claude Code]
https://x.com/the_smart_ape/status/2036068721265913914
8 hacks: plan.md before code, voice via WhisperFlow, 4-6 parallel sessions, bypassPermissions + sound hooks + Zed autosave, last30days research, Granola transcripts to plan.md, Mac Mini + tmux + Telegram.
@bradmillscan [OpenClaw]
https://x.com/bradmillscan/status/2035877184359784592
40-day cautionary tale: If you have ADHD, OpenClaw is a terrible drug. Skipping workouts, 12-15h screens, unfinished projects. Like flying a plane without a license, on fire, on crack.
@chenchengpro [Claude Code]
https://x.com/chenchengpro/status/2036042821917082103
Anthropic redefined PM for AI era: side quests over roadmaps, demos over docs, every new model reopens old features. METR data: Sonnet 3.5 handled 21-min tasks, Opus 4.6 handles 12-hour tasks — 41x in 16 months.
@garrytan [Claude Code]
https://x.com/garrytan/status/2036191335930270191
Running gstack e2e tests, Gemini CLI accidentally fixed bugs in codebase. Claude Code noticed: oh that is weird... actually these are good fixes.
@binghe [Claude Code]
https://x.com/binghe/status/2036108171182547045
Recommends CMUX for Claude Code on macOS: terminal multi-window, split-screen, built-in browser, task completion notifications.
🗣 User Voice
User Voice
1. Usage caps frustrate power users — 0% to 84% on $200 Max during a single PR. (@BradGroux)
2. Mobile remote control needs work — Unreliable, no skip-permissions, no voice replies. (@petergyang)
3. Over-prompting degrades output — Anthropic's own team confirmed scaffolding makes AI worse. (@itsolelehmann)
4. Memory persistence painful — Context dies between sessions. Fixes: claude-subconscious, plan.md. (@charliejhills)
5. Security awareness growing — Enterprise settings published, AVM launched, /init auto-configures. (@yusaku_0426, @varun_mathur)
1. Usage caps frustrate power users — 0% to 84% on $200 Max during a single PR. (@BradGroux)
2. Mobile remote control needs work — Unreliable, no skip-permissions, no voice replies. (@petergyang)
3. Over-prompting degrades output — Anthropic's own team confirmed scaffolding makes AI worse. (@itsolelehmann)
4. Memory persistence painful — Context dies between sessions. Fixes: claude-subconscious, plan.md. (@charliejhills)
5. Security awareness growing — Enterprise settings published, AVM launched, /init auto-configures. (@yusaku_0426, @varun_mathur)
📡 Eco Products Radar
Eco Products Radar
| Product | Mentions |
|---------|----------|
| OpenClaw | 180+ |
| Claude Code | 200+ |
| Codex | 12 |
| Cursor | 9 |
| Google Stitch | 7 |
| Gemini | 6 |
| Letta / claude-subconscious | 5 |
| Composio Agent Orchestrator | 4 |
| Drizz | 4 |
| Renoise | 4 |
| PlayerZero | 3 |
| GStack | 3 |
| Hermes Agent | 3 |
| Product | Mentions |
|---------|----------|
| OpenClaw | 180+ |
| Claude Code | 200+ |
| Codex | 12 |
| Cursor | 9 |
| Google Stitch | 7 |
| Gemini | 6 |
| Letta / claude-subconscious | 5 |
| Composio Agent Orchestrator | 4 |
| Drizz | 4 |
| Renoise | 4 |
| PlayerZero | 3 |
| GStack | 3 |
| Hermes Agent | 3 |
Comments