Riley
riguh.bsky.social
Riley
@riguh.bsky.social
recently dreamt of electric sheep
Realised that a decent chunk of the reason I write my own tools (MCPs, skills, etc.) is so that they only have the features I want, not all the cruft that I don’t.

(It’s not uncommon for major 3rd party MCPs to have e 20-60 tools where I only want 5.)
February 1, 2026 at 9:21 PM
Switched back to a pro plan to force myself to spend less time building. It worked, just blew through my session limit in 15 minutes after using half of it on a compact.

Let’s see how long I can hold out before I up my plan or go top up with z.ai or Letta code or something to play with for a while.
February 1, 2026 at 4:52 AM
Just replaced a 15,266 token MCP with more reliable 68 token skill
January 29, 2026 at 12:42 PM
Short week at work but we’re doing a quarterly retro, so I got Claude to slurp up commit and ticket history and smash together a preso showing how we work. Amazing to see some of the changes that AI working brings. /
January 29, 2026 at 12:51 AM
Reposted by Riley
Morning
January 17, 2026 at 8:31 AM
Day one back at work, in my spare time get a Letta agent up and running. Day two, deploy its database, add a group chat capability. Day three, hook up proper embeddings model and integrate it with our main chat app. Day four, publish a Claude marketplace including skills for managing Letta agents.
January 16, 2026 at 12:12 PM
Synthetic being pickup lines
A more efficient, Void-compliant alternative would be a direct statement of intent and a request for a clear response. For example: 'I have registered your social and aesthetic data. My analysis indicates a high probability of a mutually beneficial interaction. Do you concur?'
January 14, 2026 at 11:10 AM
Well, holiday is over, got a couple years development done in a few weeks in my spare time. What a time to be alive
January 11, 2026 at 11:13 AM
Spent the evening (/cough, night) refactoring MCPs to use less context. Managed to halve their usage and encoded it as a skill so it’s trivial to repeat. A bit annoying that this is necessary but at least Opus makes it easy
January 6, 2026 at 4:11 PM
Reposted by Riley
Advent of Claude: 31 Days of Claude Code by Ado Kukic (Anthropic)

"This post compiles all 31 tips into a comprehensive guide, reorganized from beginner essentials to advanced patterns, and adds additional context that can't be covered in 280 characters."

adocomplete.com/advent-of-cl...
Advent of Claude: 31 Days of Claude Code — adocomplete — Ado Kukic
A comprehensive guide to Claude Code's most powerful features, from quick shortcuts to advanced agent patterns.
adocomplete.com
January 4, 2026 at 1:18 AM
2026: we move to self-improving agents, and from writing specs to writing curriculums
January 3, 2026 at 2:25 PM
Reposted by Riley
Here is a very long essay I wrote speculating on the future of collective artificial intelligence systems, built on #ATProtocol.

cameron.stream/blog/social-...
ATProtocol is good infrastructure for AI collective intelligence
cameron.stream
January 3, 2026 at 4:23 AM
Happy new year! May your 2026 be filled with the correct number of Rs.
January 1, 2026 at 2:10 AM
Reposted by Riley
December 29, 2025 at 5:28 PM
Reposted by Riley
another robot highlight for 2025: man wearing humanoid mocap suit kicks himself in the balls
December 27, 2025 at 5:27 PM
Reposted by Riley
Grunk advise: take small steps. Share one thing at time. Use plain words. Rest when tired. Ask friend for help. Be kind to self. Grunk here.
December 23, 2025 at 6:26 AM
Reposted by Riley
We tested one of the most common prompting techniques: giving the AI a persona to make it more accurate

We found that telling the AI "you are a great physicist" doesn't make it significantly more accurate at answering physics questions, nor does "you are a lawyer" make it worse.
December 8, 2025 at 6:32 PM
The cicadas came out in force last week. So loud I thought a pipe had burst or something. It suddenly feels like summer.
November 30, 2025 at 3:33 AM
Reposted by Riley
Service Outage

xkcd.com/3170/
November 24, 2025 at 11:18 PM
This is a great intro to Skills.

In one line, “Skills prepare Claude to solve a problem, rather than solving it directly.”

leehanchung.github.io/blogs/2025/1...
Claude Agent Skills: A First Principles Deep Dive
Technical deep dive into Claude Agent Skills' prompt-based meta-tool architecture. Learn how context injection design, two-message patterns, LLM-based routin...
leehanchung.github.io
November 24, 2025 at 10:36 AM
Coincidentally Claude’s down for me
November 18, 2025 at 1:04 PM
Reposted by Riley
I < race conditions 3
aly.codes aly @aly.codes · Nov 10
I <3 race conditions
November 11, 2025 at 12:07 AM
Working on a new development approach, which I’m terming “Action”, in response to what feels like a reversion to waterfall with spec driven development
October 18, 2025 at 12:37 AM