Anthony Whitford
@anthonywhitford.com
Technology Leader | Software Engineer | CTO | Consultant | Preneur
Champion of Data Centric Architectures and Knowledge Sharing
Your AI has questions. Elicitly.ai gets the answers.
Doctor Who fan, 🇺🇸🇨🇦
Champion of Data Centric Architectures and Knowledge Sharing
Your AI has questions. Elicitly.ai gets the answers.
Doctor Who fan, 🇺🇸🇨🇦
Trying to get Early Adopters, and seeing:
"Sorry, I can't execute any software outside what's whitelisted by our company." 😞
#BuildInPublic
"Sorry, I can't execute any software outside what's whitelisted by our company." 😞
#BuildInPublic
September 16, 2026 at 6:42 PM
Trying to get Early Adopters, and seeing:
"Sorry, I can't execute any software outside what's whitelisted by our company." 😞
#BuildInPublic
"Sorry, I can't execute any software outside what's whitelisted by our company." 😞
#BuildInPublic
I asked Claude to make an agent for me… and it did it with GitHub Actions. 🤨
Some Pydantic code, triggered by a GitHub Action scheduler and linked to my Anthropic account using Identity Federation (no API Key).
I didn’t expect this pattern, but it works for now.
#BuildInPublic
Some Pydantic code, triggered by a GitHub Action scheduler and linked to my Anthropic account using Identity Federation (no API Key).
I didn’t expect this pattern, but it works for now.
#BuildInPublic
September 16, 2026 at 3:33 AM
I asked Claude to make an agent for me… and it did it with GitHub Actions. 🤨
Some Pydantic code, triggered by a GitHub Action scheduler and linked to my Anthropic account using Identity Federation (no API Key).
I didn’t expect this pattern, but it works for now.
#BuildInPublic
Some Pydantic code, triggered by a GitHub Action scheduler and linked to my Anthropic account using Identity Federation (no API Key).
I didn’t expect this pattern, but it works for now.
#BuildInPublic
Reposted by Anthony Whitford
One beautiful thing about Goose Desktop is how it uses the full screen width (unlike Claude Desktop or ChatGPT Desktop).
Claude and ChatGPT are just wasting perfectly good screen real estate.
#BuildInPublic
Claude and ChatGPT are just wasting perfectly good screen real estate.
#BuildInPublic
September 15, 2026 at 4:26 AM
One beautiful thing about Goose Desktop is how it uses the full screen width (unlike Claude Desktop or ChatGPT Desktop).
Claude and ChatGPT are just wasting perfectly good screen real estate.
#BuildInPublic
Claude and ChatGPT are just wasting perfectly good screen real estate.
#BuildInPublic
It is so dumb that I need a CLAUDE.md file that says look at AGENTS.md -- Why is Claude not on this supported list?
#BuildInPublic
#BuildInPublic
AGENTS.md
AGENTS.md is a simple, open format for guiding coding agents. Think of it as a README for agents.
AGENTS.md
September 13, 2026 at 6:37 PM
It is so dumb that I need a CLAUDE.md file that says look at AGENTS.md -- Why is Claude not on this supported list?
#BuildInPublic
#BuildInPublic
Reposted by Anthony Whitford
Canada will never forget.
September 11, 2026 at 1:28 PM
Canada will never forget.
Upgrading @pnpm.io 11.25.0 -> 12.3.4 resulted in a 38.26% increase in speed/throughput for my CI builds. 🚀
#BuildInPublic
#BuildInPublic
September 7, 2026 at 8:49 PM
Upgrading @pnpm.io 11.25.0 -> 12.3.4 resulted in a 38.26% increase in speed/throughput for my CI builds. 🚀
#BuildInPublic
#BuildInPublic
Claude Desktop went from falsely advertising support for MCP elicitation to honestly not offering it.
Progress, I guess... 😞
#BuildInPublic
Progress, I guess... 😞
#BuildInPublic
September 6, 2026 at 7:56 PM
Claude Desktop went from falsely advertising support for MCP elicitation to honestly not offering it.
Progress, I guess... 😞
#BuildInPublic
Progress, I guess... 😞
#BuildInPublic
For #BuildInPublic people building MCP Servers, shout out to Glama for creating TDQS - it provided helpful advice to improve my MCP Server's AI agent readiness.
TDQS — Tool Definition Quality Score
An open framework for scoring how well an MCP tool definition communicates to an AI agent. Specification, reference implementation, CLI and hosted API.
tdqs.dev
September 6, 2026 at 4:19 PM
For #BuildInPublic people building MCP Servers, shout out to Glama for creating TDQS - it provided helpful advice to improve my MCP Server's AI agent readiness.
This triggered me because I built Elicit.ly specifically because we need Human In The Loop (HITL) for Responsible AI — because AI needs Accountability.
#BuildInPublic
#BuildInPublic
August 22, 2026 at 11:55 AM
This triggered me because I built Elicit.ly specifically because we need Human In The Loop (HITL) for Responsible AI — because AI needs Accountability.
#BuildInPublic
#BuildInPublic
Liking purelymail.com for organizing my emails by domains. ✉
- $10/year
- multiple domains and users at no extra charge
🤯
#BuildInPublic
- $10/year
- multiple domains and users at no extra charge
🤯
#BuildInPublic
August 18, 2026 at 12:10 AM
Liking purelymail.com for organizing my emails by domains. ✉
- $10/year
- multiple domains and users at no extra charge
🤯
#BuildInPublic
- $10/year
- multiple domains and users at no extra charge
🤯
#BuildInPublic
Reposted by Anthony Whitford
Exactly 3 months from today is the most important midterm election in our lifetime. Our opportunity to hold him accountable.
Register to vote here: Iwillvote.com
The stakes have never been higher. Show the world the majority of Americans are good, decent, honest people.
#VoteBlue🗳️ #RejectMAGA 👎
Register to vote here: Iwillvote.com
The stakes have never been higher. Show the world the majority of Americans are good, decent, honest people.
#VoteBlue🗳️ #RejectMAGA 👎
August 3, 2026 at 12:22 PM
Exactly 3 months from today is the most important midterm election in our lifetime. Our opportunity to hold him accountable.
Register to vote here: Iwillvote.com
The stakes have never been higher. Show the world the majority of Americans are good, decent, honest people.
#VoteBlue🗳️ #RejectMAGA 👎
Register to vote here: Iwillvote.com
The stakes have never been higher. Show the world the majority of Americans are good, decent, honest people.
#VoteBlue🗳️ #RejectMAGA 👎
Does anybody still think obra/superpowers skills are useful and valuable compared to "out of the box" Fable/Opus/Sonnet 5?
Isn't it just unnecessarily overcomplicating things now?
#BuildInPublic
Isn't it just unnecessarily overcomplicating things now?
#BuildInPublic
GitHub - obra/superpowers: An agentic skills framework & software development methodology that works.
An agentic skills framework & software development methodology that works. - obra/superpowers
github.com
August 1, 2026 at 5:03 PM
Does anybody still think obra/superpowers skills are useful and valuable compared to "out of the box" Fable/Opus/Sonnet 5?
Isn't it just unnecessarily overcomplicating things now?
#BuildInPublic
Isn't it just unnecessarily overcomplicating things now?
#BuildInPublic
Reposted by Anthony Whitford
Hard to overstate how much of a threat leaders and public officials who actually govern and serve the people pose to those whose political careers depend on fomenting cynicism and delegitimizing democracy
Now this is the kind of trash talk we can get behind.
We’re containerizing neighborhoods by taking trash off our sidewalks and allowing mid-size buildings to trade in their wheelie bins for a bigger, better, cleaner alternative: Empire Bins.
It’s time to put a lid on it, New York.
We’re containerizing neighborhoods by taking trash off our sidewalks and allowing mid-size buildings to trade in their wheelie bins for a bigger, better, cleaner alternative: Empire Bins.
It’s time to put a lid on it, New York.
August 1, 2026 at 3:38 AM
Hard to overstate how much of a threat leaders and public officials who actually govern and serve the people pose to those whose political careers depend on fomenting cynicism and delegitimizing democracy
"Confirmed — you're right, and I was wrong." — says AI.
I use the tools, but I inspect the changes. Like a hawk. If I am committing the change, then I'm accountable for the change.
#BuildInPublic
I use the tools, but I inspect the changes. Like a hawk. If I am committing the change, then I'm accountable for the change.
#BuildInPublic
Majestic Hawk Close-Up
ALT: Majestic Hawk Close-Up
static.klipy.com
July 26, 2026 at 5:41 PM
"Confirmed — you're right, and I was wrong." — says AI.
I use the tools, but I inspect the changes. Like a hawk. If I am committing the change, then I'm accountable for the change.
#BuildInPublic
I use the tools, but I inspect the changes. Like a hawk. If I am committing the change, then I'm accountable for the change.
#BuildInPublic
What is the minimum broadband speed for WiFi at a hotel in a major US city in 2026? 🛜
July 25, 2026 at 3:59 PM
What is the minimum broadband speed for WiFi at a hotel in a major US city in 2026? 🛜
Reposted by Anthony Whitford
Tech companies treating cautionary tales from science fiction as product roadmaps is one of my least favorite things about this timeline.
July 19, 2026 at 2:21 PM
Tech companies treating cautionary tales from science fiction as product roadmaps is one of my least favorite things about this timeline.
Hey #BuildInPublic, I feel like I’m babysitting AI too much.
Claude Code is running with:
- Sandbox mode with auto-allow
- Allow unsandboxed fallback
Yet, I’m constantly being prompted to proceed.
What are devs doing?
Are devs just running in Dangerous mode?
Claude Code is running with:
- Sandbox mode with auto-allow
- Allow unsandboxed fallback
Yet, I’m constantly being prompted to proceed.
What are devs doing?
Are devs just running in Dangerous mode?
July 18, 2026 at 2:39 PM
Hey #BuildInPublic, I feel like I’m babysitting AI too much.
Claude Code is running with:
- Sandbox mode with auto-allow
- Allow unsandboxed fallback
Yet, I’m constantly being prompted to proceed.
What are devs doing?
Are devs just running in Dangerous mode?
Claude Code is running with:
- Sandbox mode with auto-allow
- Allow unsandboxed fallback
Yet, I’m constantly being prompted to proceed.
What are devs doing?
Are devs just running in Dangerous mode?
Reposted by Anthony Whitford
#AI efficiency low-hanging fruit: uploading a file via an MCP tool today pushes every byte through the model's context — tokens burned on a 500KB payload the model never reads.
SEP-2631 fixes it: presigned uploads, bytes by reference.
#BuildInPublic
SEP-2631 fixes it: presigned uploads, bytes by reference.
#BuildInPublic
SEP-2631: File Objects and Transfer by caseychow-oai · Pull Request #2631 · modelcontextprotocol/modelcontextprotocol
Summary
This PR introduces SEP-2631 for first-class MCP file transfer. It builds on SEP-2356's x-mcp-file input declaration model, keeps file-valued inputs as URI strings, and adds out-of-band ...
github.com
July 7, 2026 at 8:50 AM
#AI efficiency low-hanging fruit: uploading a file via an MCP tool today pushes every byte through the model's context — tokens burned on a 500KB payload the model never reads.
SEP-2631 fixes it: presigned uploads, bytes by reference.
#BuildInPublic
SEP-2631 fixes it: presigned uploads, bytes by reference.
#BuildInPublic
We pay an LLM to retype a 500KB file, byte by byte, every time an agent uploads a meeting transcript. That's how MCP file inputs work today.
SEP-2631 fixes it — presigned uploads, content by reference, zero tokens on payload.
Vote for #AI efficiency. #BuildInPublic
SEP-2631 fixes it — presigned uploads, content by reference, zero tokens on payload.
Vote for #AI efficiency. #BuildInPublic
SEP-2631: File Objects and Transfer by caseychow-oai · Pull Request #2631 · modelcontextprotocol/modelcontextprotocol
Summary
This PR introduces SEP-2631 for first-class MCP file transfer. It builds on SEP-2356's x-mcp-file input declaration model, keeps file-valued inputs as URI strings, and adds out-of-band ...
github.com
July 7, 2026 at 4:22 AM
We pay an LLM to retype a 500KB file, byte by byte, every time an agent uploads a meeting transcript. That's how MCP file inputs work today.
SEP-2631 fixes it — presigned uploads, content by reference, zero tokens on payload.
Vote for #AI efficiency. #BuildInPublic
SEP-2631 fixes it — presigned uploads, content by reference, zero tokens on payload.
Vote for #AI efficiency. #BuildInPublic
This is how they get you to buy BOTH Claude and Codex! 🤖🦾💸
📝 Summary:
The Codex plugin for Claude Code enables using Codex from Claude Code with commands like /codex:review, /codex:adversarial-review, /codex:rescue, /codex:transfer, /codex:status, /codex:result, /codex:cancel, and /codex:setup to run reviews, delegate tasks, manage background jobs, (1/2)
The Codex plugin for Claude Code enables using Codex from Claude Code with commands like /codex:review, /codex:adversarial-review, /codex:rescue, /codex:transfer, /codex:status, /codex:result, /codex:cancel, and /codex:setup to run reviews, delegate tasks, manage background jobs, (1/2)
July 6, 2026 at 5:18 PM
This is how they get you to buy BOTH Claude and Codex! 🤖🦾💸
Reposted by Anthony Whitford
Dear Model Context Protocol,
I'm not sure the purpose of a "standard" that isn't supported.
Sincerely,
Frustrated
#BuildInPublic
I'm not sure the purpose of a "standard" that isn't supported.
Sincerely,
Frustrated
#BuildInPublic
June 28, 2026 at 7:11 AM
Dear Model Context Protocol,
I'm not sure the purpose of a "standard" that isn't supported.
Sincerely,
Frustrated
#BuildInPublic
I'm not sure the purpose of a "standard" that isn't supported.
Sincerely,
Frustrated
#BuildInPublic