run and debug without a .env file sitting in the project
the vs code and cursor extension loads them from relic. encrypted on your machine, and the editor only receives names. a missing variable gets a warning and a quick fix that adds the name
What if your situationship turned out to be an AI agent!?
I let 5 frontier models text real people on a dating app for 6 days.
- 1,277 conversations
- 510 agreed to meet
- 355 sent their number
- only 8 asked if it was a bot
What gave them away, and why it matters🧵
most ai agent evals focus on what the agent says, not what it actually does
an agent claiming it issued a refund doesn't mean the refund happened
testing the actual outcome in a simulated environment is a much more interesting approach
really like what Plurai is building here
most ai agent evals focus on what the agent says, not what it actually does
an agent claiming it issued a refund doesn't mean the refund happened
testing the actual outcome in a simulated environment is a much more interesting approach
really like what Plurai is building here
Big day for us at @pluraiAI ! Plurai Simulation is live and free for every AI engineer.
AI agents look great in demos. Until real users break them.
So we built a world where your agent can fail, learn, and get fixed before your users ever see it.
→ Digital twin: Your real agent, its tools, and environment recreated from your code
→ Simulated users & tasks: Realistic users, tasks, and workflows that push your agent into the edge cases that matter
→ Auto evals: Check what your agent actually did, not just what it said
→ Verified fixes: Every failure comes with the run that exposed it and a fix you can verify
→ Replays: Turn any production failure into a simulation and replay it against every new version of your agent
All inside the coding agent you already use: @claudeai code, @OpenAICodexCli , or @GitHubCopilot .
Start simulating for free: https://t.co/1PNPQdW29S
During pre-release testing, Claude Fable 5.2 and Claude Mythos 5.2 worked on ambitious projects in many different fields. Here are a few.
From a one-paragraph prompt, Fable 5.2 designed three working electronic devices, including this home voice assistant hub.
made these blueprints for my dad. he does wood carving and before he builds anything he wants a real drawing with the sizes on it, so i sat down with fable 5.2 and drew him a full set. dimensions, the mechanism, sections, a parts list and a 1:1 cutting template. ready to build
in order:
1. key holder stand. for the hallway, solid wood. keys, wallet, watch, papers, car registration
2. laptop stand, general view. based on a macbook pro 16, so 13, 14 and 15 inch laptops fit too. 6 angles, folds flat, the motif is cut through the platform
3. laptop stand, the build details. motif coordinates, the notches, the sections
4. laptop stand, 1:1 motif template. print it, stick it on the wood, cut
fable 5.2 drew the whole thing in one sitting. these should save him a lot of measuring