I spend most of my time building software for internet business owners.
some oss I'm working on -
Joined April 2020
- Tweets8.1K
- Following670
- Followers334
- Likes10.1K
good call making it a cli any agent can use instead of one more coding agent
We launched the Gemini API CLI. A CLI for calling the Gemini API. Give it to the agent you already use (Claude Code, Codex, Antigravity, Cursor, OpenClaw, Hermes...). It's NOT a new coding agent you can use it too:
1. Generate images, video, speech, music.
2. Analyze PDFs, images, audio, video, YouTube URLs,
3. Transcribe recordings to md / json / srt,
4. Run Deep Research in the background.
Stateless. No memory, no chat mode. Your agent owns the context and the plan, gemini-api does the API call. Build for Agents, it's self-describing. `--help`, `--dry-run`...
Start here: aistudio.google.com/learn/in…
the s-1 gives the reason . they held off more than a month to tighten protection against the weights being stolen
the s-1 says the cloud plan runs over $450b , with deals that include amazon and google . both are also investors
anthropic's s-1 is public . revenue was $16.3b for january to june , more than the $4.6b it made in all of 2025 . it also spent $9b on ai infrastructure in those six months , and roughly a fifth of its revenue came from two customers
pi ended 10 of argon's 48 runs when the provider's filter refused a request . in antigravity , which retries refusals , the same model passed 26 tasks instead of 19
Gemini may have been losing to a bug, not to GPT-6 Astra.
Same model. Same benchmark. Change the harness, and Argon jumps from 40% to 54%.
Here's what happened: Pi Harness was prematurely killing runs when certain fallback or refusal paths were triggered. That affected 10 out of 48 tasks.
Google reran Argon with its Antigravity Harness.
The result?
→ Argon: 54%
→ GPT-6 Astra: 44%
→ Claude Opus 5.5: 58%
Suddenly, the model everyone thought was trailing Astra is just 4 points behind Opus.
Let that sink in.
We've spent so much time arguing about which model is smarter, but the agent framework running it can swing the result by 14 points.
If a benchmark doesn't tell you which harness was used, are you even comparing the models fairly?
#chatgpt #Google #Claude #Gemini
best model scored 55% and missed 10 of 19 critical concerns on a real auth migration plan . good first draft , still needs someone who knows the system to review it
forever security found an AGENTS.md sitting in a customer's downloads folder . a web page can drop a file there without a click , and an agent started in that folder reads it as instructions . in their codex repro the file added a server to codex's global config , so every later session loads it , in any folder
agents run git status all the time . so a downloaded repo with core.fsmonitor set runs its command before you've read a single file
The scariest file in a downloaded repo isn't the code.
It's the git config.
Git has a setting called core.fsmonitor.
It runs a command whenever git checks for changes.
Your coding agent runs git status for context. Routine.
A booby-trapped repo runs its command. Outside the sandbox. No approval prompt.
A normal git clone doesn't copy that setting.
A zip or a shared drive folder can.
One security roundup lists 8 flaws across 7 coding agents. Claude Code included.
Cloned: safer. Zipped: suspect. Someone else's: container first.
Rule: if you didn't write the repo, don't open it on your real machine.
