Pinned Tweet
The Interactions API is now generally available. 🎉 The Interactions API is the simplest way to build with Gemini for humans and agents.
- One API for Gemini models and agents.
- Antigravity Agent with a isolated remote Linux sandbox.
- Image gen with Nano Banana; music with h Omni.
- `background=True` for async, long-running interactions.
- Multimodal Tool Use & Combination.
- Dedicated skills for coding agent.
Building something new inside Google isn't easy, but it's possible. We spent almost a year on this because we think developers and agents deserve an API that is intuitive, familiar and easy to learn and use and evolves with improving capabilities and new behaviors.
Give it a try. Feature requests, bugs, complaints all to me. 👇🏻
- One API for Gemini models and agents.
- Antigravity Agent with a isolated remote Linux sandbox.
- Image gen with Nano Banana; music with h Omni.
- `background=True` for async, long-running interactions.
- Multimodal Tool Use & Combination.
- Dedicated skills for coding agent.
Building something new inside Google isn't easy, but it's possible. We spent almost a year on this because we think developers and agents deserve an API that is intuitive, familiar and easy to learn and use and evolves with improving capabilities and new behaviors.
Give it a try. Feature requests, bugs, complaints all to me. 👇🏻
28
23
275
45,648
Hey, no AI posts today. Tough day to be an engineer.
If you are using Railway, LiteLLM, PHP, GitLab, Node (brotli) please check your setup.
Below some helpful links and a prompt for your agent. Check even if you think you aren't impacted.
Don't give up. We're in this together. Help each other out.
If you are using Railway, LiteLLM, PHP, GitLab, Node (brotli) please check your setup.
Below some helpful links and a prompt for your agent. Check even if you think you aren't impacted.
Don't give up. We're in this together. Help each other out.
36
41
402
26,948
Advisories & checkers:
- Live CVE feed: github.com/advisories?que…
- PHP2Shell checker & fix: php2shell.com
- Node.js security releases: nodejs.org/en/blog/vulner… (Brotli guide: patchtheplanet.org/brotli)
- LiteLLM 1.107.0 advisory: docs.litellm.ai/blog/security-…
- Railway disclosure: blog.railway.com/p/security-dis…
- GitLab 19.5.1 patch: docs.gitlab.com/releases/patch…
- Live CVE feed: github.com/advisories?que…
- PHP2Shell checker & fix: php2shell.com
- Node.js security releases: nodejs.org/en/blog/vulner… (Brotli guide: patchtheplanet.org/brotli)
- LiteLLM 1.107.0 advisory: docs.litellm.ai/blog/security-…
- Railway disclosure: blog.railway.com/p/security-dis…
- GitLab 19.5.1 patch: docs.gitlab.com/releases/patch…
3
12
44
4,678
```
Run a read-only check on this repo, its containers, and my local environment for the Oct 21-22, 2026 security disclosures (ask before editing files):
1. Supply-chain & secrets (scanners miss these):
- LiteLLM 1.107.0: check all Python lockfiles, Docker tags (`latest`/`1.107.0`), and local `site-packages` + `/tmp` for `cook.pth`, `cook.py`, `litellm_init.pth`, `/tmp/.pc_once_*`, or `/tmp/remote_code_loader.lock`.
- Railway: check for `railway.*` configs or `RAILWAY_*` CI vars and list every secret/API key/DB URL referenced so I can rotate them at the source provider.
2. Runtimes & infra: check local versions, base images, and IaC for PHP (<8.5.12 / 8.4.27 / 8.3.36 / 8.2.35), Node/Bun Brotli decoding, self-hosted GitLab (<19.5.1, CVE-2026-110264), and eMudhra CA stores (`uv <0.11.7`, `certifi <2026.10.20`).
3. Direct & transitive deps: run available scanners (`osv-scanner`, `uvx pip-audit`, `npm/pnpm/bun audit`, `govulncheck ./...`) and check lockfiles against github.com/advisories
Return a compact table: Status (`ROTATE KEYS` / `PATCH` / `SAFE`) | Component | Location | Exact Fix Command.
```
Run a read-only check on this repo, its containers, and my local environment for the Oct 21-22, 2026 security disclosures (ask before editing files):
1. Supply-chain & secrets (scanners miss these):
- LiteLLM 1.107.0: check all Python lockfiles, Docker tags (`latest`/`1.107.0`), and local `site-packages` + `/tmp` for `cook.pth`, `cook.py`, `litellm_init.pth`, `/tmp/.pc_once_*`, or `/tmp/remote_code_loader.lock`.
- Railway: check for `railway.*` configs or `RAILWAY_*` CI vars and list every secret/API key/DB URL referenced so I can rotate them at the source provider.
2. Runtimes & infra: check local versions, base images, and IaC for PHP (<8.5.12 / 8.4.27 / 8.3.36 / 8.2.35), Node/Bun Brotli decoding, self-hosted GitLab (<19.5.1, CVE-2026-110264), and eMudhra CA stores (`uv <0.11.7`, `certifi <2026.10.20`).
3. Direct & transitive deps: run available scanners (`osv-scanner`, `uvx pip-audit`, `npm/pnpm/bun audit`, `govulncheck ./...`) and check lockfiles against github.com/advisories
Return a compact table: Status (`ROTATE KEYS` / `PATCH` / `SAFE`) | Component | Location | Exact Fix Command.
```
4
6
39
3,937
The easiest way to try Gemini 4 Argon in an agentic environment is Gemini Managed Agents. A single API call gets you
- Python, Node.js, Git, and Bash.
- Full network access and control.
- Dedicate Credentials API.
- Filesystem to persist work.
- Background Support.
W
- Python, Node.js, Git, and Bash.
- Full network access and control.
- Dedicate Credentials API.
- Filesystem to persist work.
- Background Support.
W
16
4
88
6,944
Can Agents improve their own harness on unseen tasks? The @GoogleResearch team released Harness², a framework that allows improve an harness around a frozen model.
How it works:
1️⃣ Agent runs the base harness on each task to record a baseline trajectory
2️⃣ Editor agent inspects 2 candidate edits (SI, Skills, hooks,….), and runs both.
3️⃣ Composer agent diffs the new trajectory against the baseline, merges edits that fixed failures (within a cap), and runs the final harness.
4️⃣ Reflector agent logs verified lessons into file and saves the winning components to a library for future tasks to pull on demand.
Results:
- 💡 Gemini 3.5 Flash +8.4 on WorkBuddy-Bench and +7.5 on JobBench.
- 🛠️ Skills and scripts drive ~50% of edits; dropping skills costs 5.3 pts.
- 📚 Gemini 3.5 Flash improves on all 3 benchmarks when editing its own harness.
Paper: arxiv.org/abs/2610.14224
Code: github.com/google-researc…
How it works:
1️⃣ Agent runs the base harness on each task to record a baseline trajectory
2️⃣ Editor agent inspects 2 candidate edits (SI, Skills, hooks,….), and runs both.
3️⃣ Composer agent diffs the new trajectory against the baseline, merges edits that fixed failures (within a cap), and runs the final harness.
4️⃣ Reflector agent logs verified lessons into file and saves the winning components to a library for future tasks to pull on demand.
Results:
- 💡 Gemini 3.5 Flash +8.4 on WorkBuddy-Bench and +7.5 on JobBench.
- 🛠️ Skills and scripts drive ~50% of edits; dropping skills costs 5.3 pts.
- 📚 Gemini 3.5 Flash improves on all 3 benchmarks when editing its own harness.
Paper: arxiv.org/abs/2610.14224
Code: github.com/google-researc…
53
34
325
17,234
what's your early feedback on Gemini 4 Argon? where should we improve? we take your input seriously.
1,200
40
1,379
264,427
The new Gemma docs! gemma.ai 🥁🥁🥁
From capabilities, to fine-tuning guides, inference optimizations and educational content on how the models work.
All at central place!
From capabilities, to fine-tuning guides, inference optimizations and educational content on how the models work.
All at central place!
10
8
139
8,306
Gave Gemini 4 Argon a photo of the Manhattan skyline asked it to recreate it as oil painting.
It wrote a multi-pass canvas renderer that starts with broad background strokes, blocks in the buildings and river, and finishes with fine specular highlights on the water.
It wrote a multi-pass canvas renderer that starts with broad background strokes, blocks in the buildings and river, and finishes with fine specular highlights on the water.
13
12
149
19,703
Gemini 4 Argon is now live in the Gemini API and @GoogleAIStudio 🎉 What it has done at Google so far:
- A months-long Chrome refactor at ~4x the usual pace
- 35%+ more agentic code submissions
- Improved session creation in ADK by 40%
- 2.9x the throughput increases in SGLang i 4 you should spend most of your time writing a clear goal and a way to check the result. Through your hardest challenge and convince yourself: aistudio.google.com/learn/gemini-4…
- A months-long Chrome refactor at ~4x the usual pace
- 35%+ more agentic code submissions
- Improved session creation in ADK by 40%
- 2.9x the throughput increases in SGLang i 4 you should spend most of your time writing a clear goal and a way to check the result. Through your hardest challenge and convince yourself: aistudio.google.com/learn/gemini-4…
25
13
182
10,211
Also rolling out in @antigravity (Google AI Ultra), and the @GeminiApp and AI Mode in Search (Google AI Pro and Ultra). Read more: blog.google/innovation-and…
3
0
13
1,456
Will be at @aiDotEngineer in NYC this week! 🔥
- Building a autonomous coworker | 10/14 2:55pm
- @GoogleDeepMind Booth | most of the time
Come by if you build agents, evals or have feedback about Gemini.
- Building a autonomous coworker | 10/14 2:55pm
- @GoogleDeepMind Booth | most of the time
Come by if you build agents, evals or have feedback about Gemini.
4
2
26
4,643
Add:
> can communicate technical concepts to none ai native engineers.
> pushes to keeps things simple and says no to features/behaviors
> can communicate technical concepts to none ai native engineers.
> pushes to keeps things simple and says no to features/behaviors
literal hardest role to hire for rn and every ai startup wants someone
> chronically online (knows trends)
> has taste & can create (new media)
> understands ai & technical
> can execute with quality (operator)
such a small niche of folks who can do this
8
1
28
4,549
SynthID Detector (synthid.com) is now publicly available, supporting the Nano Banana 2.1, @OpenAI, @nvidia, Kakao, and soon @Apple. 🌐
Verify if an image, video, or audio file was generated:
1️⃣ Upload or paste your file at synthid.com
2️⃣ Scans for or our partners (files are deleted right after)
Verify if an image, video, or audio file was generated:
1️⃣ Upload or paste your file at synthid.com
2️⃣ Scans for or our partners (files are deleted right after)
12
4
42
4,836
