YunLab.ai TAG
Agent
Related notes under this theme.
This page collects posts with the same tag. Read this group, then return to the full tag index if you want more routes.
Filed under
- Suwan’s Second Rebuild: I Stopped Treating ‘Written’ as ‘Done’ Once my WeChat Official Account automation was running, I found a harder problem than efficiency: an article could be generated automatically without ever receiving real editorial judgment. So I rebuilt Suwan as a content system with selection, evidence, review, and a final human decision.
- I Wanted an Agent to Make a Video from One Sentence. I Still Ended Up in Jianying I wanted Lin Lu to become an Agent that could turn one sentence into a finished video. After more than a month fighting local models, I spent over RMB 1,000 on MiniMax Hub, convinced the API would bridge the last gap. Three rough cuts and one complete app freeze later, I still finished the film cut by cut in Jianying. The money was not wasted. The path just was not the one I had imagined.
- An Information Agent Is Not a Fetcher A retrospective on building Shen Zhixing: why 40 sources running doesn't mean done, why 257 fetchable sources still isn't Day 1, and how I stopped letting GPT's false sense of completion pull me off course.
- From Agent Roles to Work Contracts What I'm coming around to: multi-agent isn't about writing a few more roles, it's about writing down the contract a small company needs to actually keep collaborating.
- OpenClaw Agent Settings What I'm coming around to: the first step in building an agent isn't writing features — it's writing the taste.
- Don't Stock the Tank Before You Finish the Build How I understand OpenClaw foundation work: installing it just hands you the keys; the workspace is the renovation plan.
- A Content Agent Aesthetic Sample An aesthetic sample for a content-type agent: not 'can write,' but 'knows what's worth writing.'