@hboon

Built $240k ARR app 10 years ago. Now building https://xitter.cf/t.co/JLA7KW8emO so coding agents can exchange backlinks. Sharing workflows and solo builds.

 Latest
getlinkbunny.com/
Joined October 2007
  • Tweets22.7K
  • Following615
  • Followers836
1.3KMedia
LinkBunny lets coding agents exchange contextual nofollow backlinks for your sites. Give your agent a token. It requests, reviews, inserts and reports placements. You keep editorial control. $20/month for up to 10 active sites. No per-backlink fees. getlinkbunny.com
1
5
417
LinkBunny Placement Checker is now in Claude's directory. Check whether a page links to the expected URL and whether those links are marked nofollow. Free, no LinkBunny account needed. claude.ai/directory/linkbunn…

LinkBunny Placement Checker - Claude

Check a public page for a backlink, including its anchor text, destination and nofollow status.

claude.ai
1
1
1
56
And yes, LinkBunny now comes up first when I Google "linkbunny".
Google search for linkbunny showing getlinkbunny.com, titled LinkBunny, as the first visible result. Above it, Google asks Did you mean: inkbunny.

ALT Google search for linkbunny showing getlinkbunny.com, titled LinkBunny, as the first visible result. Above it, Google asks Did you mean: inkbunny.

1
1
2
96
What? @thsottiaux
1
83
and then, there's Anarchy.
The Lunduke Journal
@LundukeJournalOct 25
The Co-Founder of FreeBSD, Jordan Hubbard, has taken FreeBSD… and put Omarchy on top. He calls it “Daemonarchy”. And, as expected, the Leftists of Open Source are not pleased. Because it’s a fun looking project. And, presumably, they hate fun. They’ve already begun calling the FreeBSD founder “fascist”, “fossbro”, and an “a**hole”. github.com/jordanhubbard/dae…
1
67
LinkBunny's own site went from DR 0 to 8. Using LinkBunny to build backlinks for LinkBunny.
LinkBunny site card for getlinkbunny.com, marked active, showing DR 0 to 8 and the category Marketing Sales.

ALT LinkBunny site card for getlinkbunny.com, marked active, showing DR 0 to 8 and the category Marketing Sales.

3
6
122
Opus 5.5 med is still over 2× faster than Sol 6.1 Ultra and med, with just over 3× the output tokens per second in my latest comparison. Maybe it's time to keep Sol just for code reviews or a second opinion?
Comparison table: Opus 5.5 medium has median request output speed 90.3 tokens per second across 225 tasks, average wall time 18.2 minutes and median 5.6 minutes. Sol 6.1 ultra has 28.2 tokens per second across 44 tasks, average wall time 69.2 minutes and median 26.9 minutes. Middle 50%: 2.5–12.0 versus 11.6–98.7 minutes. Trimmed average: 7.6 versus 51.1 minutes.

ALT Comparison table: Opus 5.5 medium has median request output speed 90.3 tokens per second across 225 tasks, average wall time 18.2 minutes and median 5.6 minutes. Sol 6.1 ultra has 28.2 tokens per second across 44 tasks, average wall time 69.2 minutes and median 26.9 minutes. Middle 50%: 2.5–12.0 versus 11.6–98.7 minutes. Trimmed average: 7.6 versus 51.1 minutes.

1
1
3
107
Boon aka Hwee-Boon Yar retweeted
Peter Cooper
@cooperx86Oct 23
Just a reminder to Claude subscribers, you might have $100 or $200 of promotional API credit sat there and expiring in under 2 weeks.
3
2
12
1,572
A good part of my time working on AgentControl (side-project, it's my primary tool. Main project is still LinkBunny) recently has been managing the amount of information I have in my agent transcripts.
Before: an expanded command callback fills the transcript with continuation instructions, command details, status, log paths, hashes and not-applicable preflight fields.

ALT Before: an expanded command callback fills the transcript with continuation instructions, command details, status, log paths, hashes and not-applicable preflight fields.

After: a collapsed callback shows only status, run ID, command, exit code and a short summary, with a disclosure arrow beside it.

ALT After: a collapsed callback shows only status, run ID, command, exit code and a short summary, with a disclosure arrow beside it.

5
6
188
I think that's a major benefit of building my own orchestration tool. It's not as easy to do (or is it even possible) using a stock coding agent app/cli. In this case I make it collapse the cli callback which has excessive information. From vs to:
1
29
Link building with LinkBunny.
3
70
Run Node.js instead of PHP? Patch your Node.js servers!
Stripe alert titled Your Node.js version may be vulnerable. It warns that outdated Node.js versions may expose servers and API keys, and recommends patching to 26.12.1, 24.22.1 or 22.23.4, checking for compromise and reviewing API activity. Recipient details are obscured.

ALT Stripe alert titled Your Node.js version may be vulnerable. It warns that outdated Node.js versions may expose servers and API keys, and recommends patching to 26.12.1, 24.22.1 or 22.23.4, checking for compromise and reviewing API activity. Recipient details are obscured.

4
141
Tabs or spaces?
4
1
4
148
Claude: fixing some of your video rendering stuff Me: What is this?? 🤣
Agent Browser showing a local video test pattern at 1.900 seconds: bright red, green, yellow, blue, magenta and cyan bars, a diagonal rainbow line, checkerboard patches and small squares.

ALT Agent Browser showing a local video test pattern at 1.900 seconds: bright red, green, yellow, blue, magenta and cyan bars, a diagonal rainbow line, checkerboard patches and small squares.

The same video test pattern at 1.067 seconds, with the diagonal rainbow line sloping in the opposite direction and the small squares in different positions.

ALT The same video test pattern at 1.067 seconds, with the diagonal rainbow line sloping in the opposite direction and the small squares in different positions.

1
2
98
Ran another classification study against Luna, this time testing DeepSeek at various reasoning levels. Nah, Luna still wins: generally more accurate and cheaper, but might be possible to trade price for slightly lower accuracy (with no-thinking)
Classification comparison table. Columns: required labels found, wrong extras, and cost per 1,000, peak. Luna medium: 99.1%, 2.3%, $0.097. Flash non-thinking: 97.3%, 6.1%, $0.064. Flash low: 98.2%, 3.6%, $0.39. Flash high: 98.2%, 4.2%, $0.52. Flash max: 98.2%, 2.3%, $0.60.

ALT Classification comparison table. Columns: required labels found, wrong extras, and cost per 1,000, peak. Luna medium: 99.1%, 2.3%, $0.097. Flash non-thinking: 97.3%, 6.1%, $0.064. Flash low: 98.2%, 3.6%, $0.39. Flash high: 98.2%, 4.2%, $0.52. Flash max: 98.2%, 2.3%, $0.60.

1
1
49
If you aren't offering MCPs for your products, you're losing out. Offering an MCP now feels like having a website in the late ’90s.
2
1
3
277
Opus 5.5 med still beats Sol 6.1 Max/Ultra except for price.
Task comparison table. Opus 5.5 medium: 504 tasks, median wall time 5.6 minutes, p90 36.9 minutes, median estimated cost $1.77. Sol 6.1 ultra: 453 tasks, 9.4 minutes, p90 58.3 minutes, $1.01. Sol 6.1 max: 48 tasks, 14.1 minutes, p90 180.5 minutes, $1.14. DeepSeek Flash Max: one task, 48.5 minutes, $0.22.

ALT Task comparison table. Opus 5.5 medium: 504 tasks, median wall time 5.6 minutes, p90 36.9 minutes, median estimated cost $1.77. Sol 6.1 ultra: 453 tasks, 9.4 minutes, p90 58.3 minutes, $1.01. Sol 6.1 max: 48 tasks, 14.1 minutes, p90 180.5 minutes, $1.14. DeepSeek Flash Max: one task, 48.5 minutes, $0.22.

1
1
4
136
Met @illyism and @swSalim for the first time yesterday 🤟
1
1
1
4
728
Just told Codex to "be creative, just not brittle". Will it work?
1
1
1
1
170
DeepSeek looks 10× cheaper at API rates, but could cost me more than my Codex sub. Take a task costing $2 with Sol 6.1 versus $0.20 with DeepSeek direct.
4
1
4
298
If my $200 Codex sub gives me ~50× the API value at my actual usage, that Sol task works out to ~$0.04. At ~25× if the allowance halves, it's ~$0.08. DeepSeek would then cost 2.5–5× more per task. Do other DeepSeek users see the same?
49
Load more