//
sign in
Profile
by @danabra.mov
Profile
by @dansshadow.bsky.social
Profile
by @jimpick.com
AviHandle
by @danabra.mov
AviHandle
by @dansshadow.bsky.social
AviHandle
by @katherine.computer
EventsList
by @katherine.computer
ProfileHeader
by @dansshadow.bsky.social
ProfileHeader
by @danabra.mov
ProfileMedia
by @danabra.mov
ProfilePlays
by @danabra.mov
ProfilePosts
by @danabra.mov
ProfilePosts
by @dansshadow.bsky.social
ProfileReplies
by @danabra.mov
Record
by @atsui.org
Skircle
by @danabra.mov
StreamPlacePlaylist
by @katherine.computer
+ new component
ProfilePosts









Loading...
Concise Guide to Writing Effective Agent Skills (from SkillsBench paper) Define procedural focus: - Deliver how-to guidance only – workflows, standard operating procedures (SOPs), domain conventions, and heuristics for a class of similar tasks (not factual recall or single-instance solutions).
- Include exact API calls, function names, parameters, step-by-step sequences, output-format reminders, and at least one concrete working example. Ensure reusability and portability:
Use required structure: - Place everything in a modular directory containing SKILL md (natural-language instructions with YAML frontmatter for name and description) plus optional resources (code templates, executable scripts, reference docs, or worked examples). Keep focused and concise:
- Skills must be accurate, internally consistent, clear, specific, and error-free; expert-written skills drive the percentage gains while self-generated ones deliver no benefit.
- Limit to 2–3 core modules or procedures per skill/task; detailed or compact guidance outperforms exhaustive/comprehensive documentation. Make it actionable:
- Write for file-system use across agents; avoid any task-specific leakage (no test-case constants, filenames, or paths). Prioritise human curation and quality:
Is GPT-5.6 to arrive with stronger safety checks? The below message appeared from a discussion in ChatGPT saying it required more time to check. x.com/i/status/20...
The real question is whether they’ll actually make it public. 🤔 x.com/i/status/20...
1h
1h
1h
1h
1h
2h
1h
3h
Sounds like GPT-5.6 is going to be much better at front-end development. It's about time. I'd love to see a Codex Design tool just like we have Claude Design. 5.6 and 5.6 Pro are rumoured to be next Thursday atm. Could we also get Sonnet 5 and Gemini 3.5 Pro? x.com/i/status/20...
Kol Tregaskes
Kol Tregaskes
Kol Tregaskes
Kol Tregaskes
38m
Kol Tregaskes
Kol Tregaskes
Kol Tregaskes
Kol Tregaskes
Kol Tregaskes
3h
An updated Claude Mythos is on the way. No surprise - Anthropic keeps iterating, just as it does with the rest of the lineup. Odds are it’ll be Mythos or Fable 5.1. A jump to 6 only really makes sense if Sonnet or Opus also move to version 6.
Kol Tregaskes