robotoole | stocktools.app

arrest this guy wtf is this shit

days like today, my premium engine strategy keeps me feeling somewhat comfortable. given the married put position is easy to maneuver in these down trends. when i started these positions, i bought a 55 delta long put ~120dte and then sold calls against the position. times like these when we're way down, i like to roll in the put, get it exercised so all my shares get called away above my cost basis. i'm flat to up in all these positions because i've been selling calls and puts the entire time making 1-2% a week. if only i didn't also have some leaps and short puts to go with the shares that are also in trouble. it's not all roses when you're trying to collect premium every week. i can manage the puts easily. the few leap calls i'm holding on the other hand... those deltas are dying on the vine. i'll be happy to have the cash/buying power back on monday from my shares. hope we get some peace soon. stocktools.app

@pbakaus it's really hard to remember all the commands and skills. is there a simple catch all /impeccable or /design-review skill that will incorporate the whole thing?

Impeccable v1.6 is out. /critique added Nielsen's 10 heuristics, tests w/ persona archetypes, and assesses cognitive load. /audit scores 5 tech quality dimensions w/ severity ratings. + Trae support, better skill descriptions for agent auto-discovery. impeccable.style

@noahzweben I loop an on duty engineer that checks for new bugs in sentry and auto triages and fixes bugs. Then it keeps all open branches up to date with main. Then reviews server performance. Works Well! Need to try the new server tasks next.

Loops now run for up to 7 days instead of 3. Let me know what you’re /looping on!

@danshipper reminds me of @rands stables vs volatiles. the volatiles are having a blast and the stables have more to clean up than ever. win/win - randsinrepose.com/archives/stabl…

new model for engineering team structure in 2026: 2 people only one pirate and one architect the pirate's job is to move as fast as possible to develop valuable, shipped product features by vibe coding. the architect's job is to turn the product surface discovered by the pirate into a reliable, structured machine—also by vibe coding, but at a slower, more well-reasoned pace. every product needs a pirate but most product's only need an architect once they some form of PMF, and in that case they usually don't need one full-time. architects can work across many codebases and solve interesting technical challenges. pirates go hard on a product that they own end-to-end.

@ctatedev Switching to json-render + yaml greatly improved my ux. Thanks for the great tools!

New json-render example: game-engine AI → JSON → Render → Game - Rapier physics and gravity - Import models and sound fx - Health, take damage, build levels - First-person WASD + pointer controls We may never get GTA 6 But at least we have Generative Gaming

@euboid That's better than what I have setup. I have a Claude loop for my on duty engineer skill that checks sentry for any new crashes. Then it checks all prs and looks for merge conflicts and out of date branches with main.

This is the best thing I've ever built. When Sentry detects a crash, it sends a webhook to codex to review. Codex has access to our logs, traces and code. It triages, and creates a PR, or closes the sentry issue if it was noise. Then I merge in one click.

@sterlingcrispin i do the same but also add in cursor's bugbot as well. they both find bugs claude doesn't. claude still builds the original system better than both of them though. go figure. :D

Claude 4.6 is a good programmer but writes insanely severe bugs constantly, it won't catch them all in audits, nor will other claudes You need codex 5.4 auditing every commit 4+ times. If you don't believe me, try it. I have an /auditcodex skill for it github.com/sterlingcrispi…

@DylanCaponi was searching to see if someone else was getting as frustrated as i was with the compound commands. thanks!

claude loves to write compound bash commands. This parses and vets each one individually to auto-accept or prompt the user

An approach to not have to babysit claude code agents: github.com/dylancaponi/cl…

@startupideaspod i build ui's in my app to test the api's easier than going through chat, then expose everything that the dashboards/ui can do to the chat via mcp so the users can talk to ai/agents runs instead of doing it themselves.

The old SaaS playbook is dead. Before: Great UI = great product. API was a nice-to-have. Now: Agents run 24/7 They live in terminals They talk through MCPs The UI is the nice-to-have. The API is the product. Sam Altman said it best: Every company is going to be an API company. If your API can't do what your UI does, you're going to lose customers.

Did you know that RODY has friends? One of his friends is Theta Fortress aka T-Fort. This friend has some serious security that will back you up while you collect that sweet ass income For RODY, T-Fort, or other strategies feel free to check out @robotoole website See comments

@RoundtableSpace Moon dev has been grinding for a while. Glad to see him getting some headlines

THIS GUY IS TEACHING OPENCLAW HOW TO BUILD TRADING BOTS BY WALKING THROUGH THE ENTIRE SYSTEM ONCE. THEN DISTRIBUTING THAT EXACT TRAINING TO EVERY AGENT.

@gregisenberg What are the best marketing skills for Claude code?

you’re vibe coding when you should be vibe marketing (with claude code, openclaw etc) 1. opus 4.6 / codex 5.3 → ship the product (core features, backend, auth, infra ) 2. claude code → design the playbook (content formats, hook templates, lead magnets, reply rules, tone of voice, weekly experiments) 3. openclaw → run the playbook 24/7 (ads, post drafts, build free tools, repurpose content, create campaigns, reply to comments + dms, send follow-ups, queue experiments) 4. dashboards → decide what to double down on (saves, shares, replies, clicks, signups) most founders stop at step 1. "wHy tHis DiDn'T woRk" the money/opportunity is steps 2–5. media is the most mispriced asset right now. it isn't code, you know this relax on the vibe coding for 1 sec vibe marketing is your friend it answers the real question nobody wants to ask: “does anyone notice, remember, or care?”

@Yuchenj_UW They still use copilot lol

How did Anthropic ship this before Microsoft 365 Copilot?

@AlexFinn An engineer on my team does that at the end of the story as well. Makes the code writer write an audit for the other llm to review the output. I'm always amazed when bugbot never finds any bugs in his pr's.

This is the best AI coding setup ever: Codex 5.2 xhigh on the left, Claude Code Opus 4.5 on the right 1. Start building with Claude Code in plan mode 2. Copy and paste every plan into Codex. Validate the plan. If they have any critical feedback, give it back to CC 3. Any time CC messes up, go back to Codex and have it fix the issue 4. Every change either agent makes, ask the other agent to review it 5. Primarily use whichever agent that is the last one to clean up the other's mess Not only will you produce apps 10x faster, you'll have fun just feeling like you're managing a team of engineers Trust me, the results are worth the price you pay

@gregisenberg So many new apps need marketing help. Mine included

startup idea for you: start an AI ads agency someone will make $50M+ in this space

@burkov It's way more cost effective to have multiple max accounts. Figured that out the hard(expensive) way

I tested Claude Code with an API key where you pay for tokens. Every code update request for my apps costs about $0.80 in token cost. I make a couple hundred requests every day with my Max subscription ($100/month). If I paid for tokens, that would cost me around $80/day. So, Claude sells Max about 30 times cheaper than it sells Opus to third parties. There's no way a third-party IDE company without its own Claude Opus-like model can provide a competing agentic coding service to Anthropic. Cursor? Are you kidding me?

@claudeai Add PreToolUse hooks to whitelist only your MCP tools. Block Bash, Read, Write, Edit, Glob, Grep in PreToolUse hooks. Default permissions are WIDE open. @bcherny

be very careful with the @claudeai agent sdk in production. if you're not actively blocking bash, grep, glob, etc it can and will edit and update your own codebase in production if asked nicely.

@trq212 Ty vvvm

@robotoole already in

We're rolling out Tool Search in Claude Code to reduce how much context MCP servers take up.