Building @composio | MIT PhD | IITB
San Francisco, CA
- Tweets 438
- Following 298
- Followers 426
- Likes 1,205
Vin Agarwal retweeted
We have 150k+ developers on @composio every day, and they/their agents keep asking for more tools across domain.
If you’ve built one, reply here and let’s get it listed. If you want to charge for it, Composio Instant handles billing too.
1
3
11
6,483
Vin Agarwal retweeted
Hiring is no longer a problem, We built the best Recruiter agent
It finds the people you want to hire and gets their email for a few cents.
Here's how it works:
- Describe the hire: "Founding eng, NYC, fintech"
- The agent searches for free, and Jev scores every candidate against your must-haves
- Reveal any email for cents, and see the exact price before you pay
Powered by @googleaidevs Gemini 4 Argon @typesafeai @composio
Fully free and open source: github.com/composio-communit…
5
6
35
2,604
Vin Agarwal retweeted
Huge 🚀 🚀 🚀
The largest math proof in human history is now computer-verified.
Today, Axiom announces a complete Lean auto-formalization of the Classification of Finite Simple Groups, the "periodic table of symmetry."
About 15,000 pages, built by over 100 mathematicians across generations.
3
3
25
2,514
Composio product velocity 🚀
Agents have a trust problem.
Today’s models can do most work. What they need is the right boundaries. In traditional software, those were set at build time. With agents, they have to be set and enforced at runtime.
Today we’re introducing Composio Firewall, the runtime boundary for agents.
Millions of users and developers already use Composio as the runtime for their agents - giving agents tools, auth, and sandboxed execution. Now with our agent firewall every action an agent takes is evaluated using Jev and custom SLMs tuned by Composio based on real-world usage data.
The firewall strengthens that existing boundary in both directions, checking the actions agents take and the information coming back to them through Composio.
The advantage of the runtime boundary is that Composio Firewall verifies that the agent respects user intent and the boundaries set by the developer and user. Additionally, this provides the safety layer required to protect the user against malicious prompt injection as well as other attacks.
Composio Firewall is live now for all Composio developers.
Check it out at composio.dev/firewall
1
4
147
Vin Agarwal retweeted
Agents have a trust problem.
Today’s models can do most work. What they need is the right boundaries. In traditional software, those were set at build time. With agents, they have to be set and enforced at runtime.
Today we’re introducing Composio Firewall, the runtime boundary for agents.
Millions of users and developers already use Composio as the runtime for their agents - giving agents tools, auth, and sandboxed execution. Now with our agent firewall every action an agent takes is evaluated using Jev and custom SLMs tuned by Composio based on real-world usage data.
The firewall strengthens that existing boundary in both directions, checking the actions agents take and the information coming back to them through Composio.
The advantage of the runtime boundary is that Composio Firewall verifies that the agent respects user intent and the boundaries set by the developer and user. Additionally, this provides the safety layer required to protect the user against malicious prompt injection as well as other attacks.
Composio Firewall is live now for all Composio developers.
Check it out at composio.dev/firewall
30
31
118
44,510
Powered by @composio ! Lovely working with @NousResearch
7
217
Vin Agarwal retweeted
It's time to connect 🌐
connectors for hermes agent!
- launching today w/ 50+ connectors.
- available to everyone (no account needed!)
- link your nous portal account to auto-sync all connectors and settings.
- connect and alias multiple accounts.
- limit a connector based on read/write/destructive action.
- RBAC and policies for teams on nous portal.
coming soon:
- more connectors!
- bring your own MCPs and sync them.
- select tool access during authentication.
43
55
1,157
77,511
Vin Agarwal retweeted
composio is really becoming the one stop shop for agents
> pick your harness
> connect it to composio
bam you just gave your agent every tool, app connector, skill, and more
and it's literally just with a single MCP
love this company
We’re giving away $1 million to agents.
Introducing Composio Instant: the easiest way for agents to pay for their own tools.
Claim your agent's share: composio.dev/instant
10
5
99
8,370
connectors for hermes agent!
- launching today w/ 50+ connectors.
- available to everyone (no account needed!)
- link your nous portal account to auto-sync all connectors and settings.
- connect and alias multiple accounts.
- limit a connector based on read/write/destructive action.
- RBAC and policies for teams on nous portal.
coming soon:
- more connectors!
- bring your own MCPs and sync them.
- select tool access during authentication.
32
23
461
97,452
LFG!!🚀
We’re giving away $1 million to agents.
Introducing Composio Instant: the easiest way for agents to pay for their own tools.
Claim your agent's share: composio.dev/instant
1
5
144
Vin Agarwal retweeted
being working on Composio's search tool for the last month or so and made
a couple of claude mods using @composio cli and video using @fframes_rust its so good
7
8
29
3,002
Vin Agarwal retweeted
We are launching Open Dots, run the same capabilities in 1/10th of the cost
OpenAI shipped Dot yesterday for Pro users.
Ours is open source, runs on your Mac and cheaper
Here's how it works:
- each dot has its own browser that stays logged in securely
- it can use and setup triggers on your Gmail, Calendar and 1,500 other apps
- you can call it and give it things to do while you talk
- dots can coordinate and schedule reminders
built with @composio @OpenRouter
free and open source: github.com/composio-communit…
175
214
2,380
239,489
Good news: it already exists. It’s called Composio :)
OpenRouter: 500+ models
Composio: 1,384 apps (and growing)
Even better news: There is a generous free tier with 100K tool calls/month.
20
18
428
40,280
Vin Agarwal retweeted
Tomorrow's conversation is a special one for me.
We chatted with @Jainarvind, founder of @Glean, the person whose company gave me my first job in tech.
He came with takes: why your enterprise agents behave like day-one employees, why open source just became production-ready, and why counting tokens will soon look as strange as counting CPU cycles.
Full episode tomorrow.
6
8
58
8,323
Vin Agarwal retweeted
32,264 stars on scientific-agent-skills. 38,000+ across everything we've open sourced. Behind every one of those is a scientist who tried the tools, found the rough edges, and told us about them. We're building K-Dense for you. A lot more coming this year.
5
11
888
Vin Agarwal retweeted
Always fun to chat with @pk_iv!!
"Every enterprise has realized they have to make their UX agentic, or the current one will die."
We sat down with @KaranVaidya6, Co-Founder of @Composio, at our @aiDotEngineer booth.
- Why 2026 is the year agents actually work
- What actually counts as an "agentic harness"
- Distilling 10M+ daily tool calls into reusable skills
- How self-improvement becomes the next milestone
1
2
16
2,442
We ran GPT-5.6 Sol and Claude Fable 5 head to head on 47 real agentic SaaS tasks.
One behavior stood out: Claude Fable 5 looks twice before acting. GPT-5.6 Sol is a straight shooter.
The result: fewer tool calls (see below) whenever a task has an obvious short path. Sol found that path and took it, skipping the schema lookups and double-checking.
The catch: Sol failed 2 tasks (45/47), largely due to that same straight-shooter tendency. Fable's double-checking earned it a perfect 47/47 on our benchmark.
The lesson: Model choice shouldn't be black and white. Benchmark your own workloads before you commit to a single model.
15
7
81
8,143