tweetindex
EN

Monk Zero ✓

@NoCommas

@antigma_labs, prev: @awsCloud, @Meta, @Mysten_Labs. A Turing Complete mind, wandering the world of Gödel Incompleteness.

1 690Followers
997Following
2 245Posts total
1.5KViews on collected posts

Latest posts

@NoCommas The idea of separating the agent engine from the interface makes a lot of sense. 69 views · 5 likes · 0 reposts · 1 replies Open on X →
Read the full architecture essay: https://t.co/Yg9HLhJ1zN Explore the code & build your own harness: https://t.co/IMhXSNLXZL Protocol spec: https://t.co/SWmqKOZSMH 123 views · 8 likes · 0 reposts · 1 replies Open on X →
1. Build your own coding agent harness (`mini-tui`) Because the UI belongs to you, Ante's `mini-tui` is a one-file Rust client: streaming tokens, inline tool lifecycle, interactive y/n approvals, and Esc interrupt. https://t.co/fivNuHCU42 148 views · 10 likes · 0 reposts · 2 replies Open on X →
3. Build your own claw-like personal assistant (`ante-gateway`) Run Ante as a multi-channel bot in Slack and Discord over a shared Unix socket. Every thread or DM gets an isolated session. In-chat approvals (yes/no), task stops, and workspace controls. Your team owns chat; Ante 134 views · 8 likes · 0 reposts · 1 replies Open on X →
2. Integrate with your IDE or orchestrator (`ante-acp`) Editors like Zed and JetBrains talk Agent Client Protocol (ACP). `ante-acp` maps ACP requests directly to Ante's engine over stdio. Live diffs and native permission dialogs—without modifying the core engine (WIP in repo). 126 views · 8 likes · 0 reposts · 1 replies Open on X →
`ante serve`: we are literally there to serve. Ante runs headless across stdio, local Unix domain sockets, and WebSockets: • Operations go in (prompts, tool approvals, interrupts) • Events come out (token deltas, tool activity, diff previews) 392 views · 8 likes · 0 reposts · 1 replies Open on X →
The UI and prompts belong to the user. Most coding agents lock you into their terminal, their window, or their proprietary IDE. They treat the interface as their moat. We take the opposite stance: we are there to serve. 346 views · 11 likes · 0 reposts · 1 replies Open on X →
Plugins at the surface. Streams at the seams. UI and prompts are where the user is the oracle: instant feedback, infinite malleability. The agent core is brutal to verify: eval sweeps burning billions of tokens. Don't fuse them together. Plugin the surface. Protect the 211 views · 7 likes · 0 reposts · 1 replies Open on X →

Against accounts of the same size

8 posts from the last 90 days, next to the under 10K follower range. reaches fewer people than peers, but engages them much harder.

Median views141this account4 567median for under 10K
Reach, %8.34%this account271.32%median for under 10K
Engagement, %6.93%this account1.41%median for under 10K
MetricThis accountMedian for under 10KRatio
Median views per post1414 5670.03×
Reach (views ÷ followers)8.34%2.7× audience0.03×
Engagement rate6.93%1.41%4.92×

Others in this range →   Compare with another account →   How these benchmarks are built →

Growth & engagement

How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.

Views per post

21121 Sep
346
392
126
134
148
123
69

Last 8 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.

Engagement rate per post

3.79%21 Sep
3.47%
2.55%
7.14%
6.72%
8.11%
7.32%
8.70%

Reactions — likes, reposts, replies and quotes — divided by views. Median for under 10K accounts is 1.41%.

What the audience does

Likes77.4%65 in total
Replies10.7%9 in total
Quotes1.2%1 in total
Bookmarks10.7%9 in total

Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.

The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.

Similar accounts