Hey @jasontheadams @WordPress Since MCP Adapter 0.7.0, requests in a 2025-06-18 or 2025-11-25 session without an MCP-Protocol-Version header get HTTP 400. 0.6.x accepted them, and the 2025-11-25 spec lets a server use the version negotiated at initialization instead: modelcontextprotocol.io/spec…
Could the adapter fall back to the session's version, or add a filter, so clients or proxies that drop the header don't break?
Also: posting a JSON-RPC response object logs PHP Warning: Undefined array key "method" (McpWireOrchestrator.php:409).
1
1
137
Hey @mattyp @bot bug report for Grok Bot's MCP connector: tool calls arrive without the MCP-Protocol-Version header (the session ID is sent). The MCP spec says clients MUST send it on every request after initialize, and WordPress MCP Adapter 0.7.0 rejects calls without it (HTTP 400).
Spotted this when connecting Grok Bot to WordPress sites via my plugin that was developed to provide a governance layer for AI agents to act on sites under human-approved rules. I've updated my plugin to address this issue, but ideally it should be addressed on your end 🫶
I think there's a huge market opportunity in building software that acts as a governance layer for AI agents.
Many organisations are waiting for robust guard rails that give them confidence to give AI agents access their systems.
The fear of agents accidentally breaking things, destroying data or causing security issues is very real, particularly for SMEs.
I had this same fear, so I developed a governance layer for all of our sites running on @WordPress. It allows me to use agents from any of the frontier AI labs with confidence.
The plugin provides a governance layer that lets AI agents act on the site under human-approved rules. It has policies for each action (with allow, require approval, or deny), an approval inbox, email and/or @SlackHQ notifications, changes tracking, audit trail, kill switch, and governed MCP connections. Agents can create posts, build pages, make design changes, manage products, and more.
I built the plugin in my spare time over a couple of weeks, including time for extensive testing and numerous rounds of auditing.
It works so well that it feels like magic.
About 10 years ago, I was developing and selling WordPress plugins. I've been building a lot again recently, and to say that my development capabilities have been supercharged by AI is an understatement.
What a time to be alive.
179
Replying to @NotTomBrown
Same here.
By way of background for those who care, I spent a lot of time last week with senior members of the Anthropic team to understand what they do to ensure Claude is good for humanity and was impressed.
Everyone I met was highly competent and cared a great deal about doing the right thing. No one set off my evil detector. So long as they engage in critical self-examination, Claude will probably be good.
After that, I was ok leasing Colossus 1 to Anthropic, as SpaceXAI had already moved training to Colossus 2.
The Robot Suit retweeted
when Claude Opus 6 tells you to "stop spiraling and go to bed" 😵💫
🤖 Made with AI
The Robot Suit retweeted
Adopting Claude speak in my regular life, episode 1:
Partner: Did you do the dishes tonight?
Me: Yes they're done.
Partner: Why are they still dirty?
Me: You're right to push back. I didn't actually do them.
Closer to being true than most people realise. There are multiple paths to dramatically increasing human lifespan, both biological and cybernetic. @bryan_johnson is onto something.
3
186
The Robot Suit retweeted
Judging by my tl there is a growing gap in understanding of AI capability.
The first issue I think is around recency and tier of use. I think a lot of people tried the free tier of ChatGPT somewhere last year and allowed it to inform their views on AI a little too much. This is a group of reactions laughing at various quirks of the models, hallucinations, etc. Yes I also saw the viral videos of OpenAI's Advanced Voice mode fumbling simple queries like "should I drive or walk to the carwash". The thing is that these free and old/deprecated models don't reflect the capability in the latest round of state of the art agentic models of this year, especially OpenAI Codex and Claude Code.
But that brings me to the second issue. Even if people paid $200/month to use the state of the art models, a lot of the capabilities are relatively "peaky" in highly technical areas. Typical queries around search, writing, advice, etc. are *not* the domain that has made the most noticeable and dramatic strides in capability. Partly, this is due to the technical details of reinforcement learning and its use of verifiable rewards. But partly, it's also because these use cases are not sufficiently prioritized by the companies in their hillclimbing because they don't lead to as much $$$ value. The goldmines are elsewhere, and the focus comes along.
So that brings me to the second group of people, who *both* 1) pay for and use the state of the art frontier agentic models (OpenAI Codex / Claude Code) and 2) do so professionally in technical domains like programming, math and research. This group of people is subject to the highest amount of "AI Psychosis" because the recent improvements in these domains as of this year have been nothing short of staggering. When you hand a computer terminal to one of these models, you can now watch them melt programming problems that you'd normally expect to take days/weeks of work. It's this second group of people that assigns a much greater gravity to the capabilities, their slope, and various cyber-related repercussions.
TLDR the people in these two groups are speaking past each other. It really is simultaneously the case that OpenAI's free and I think slightly orphaned (?) "Advanced Voice Mode" will fumble the dumbest questions in your Instagram's reels and *at the same time*, OpenAI's highest-tier and paid Codex model will go off for 1 hour to coherently restructure an entire code base, or find and exploit vulnerabilities in computer systems. This part really works and has made dramatic strides because 2 properties: 1) these domains offer explicit reward functions that are verifiable meaning they are easily amenable to reinforcement learning training (e.g. unit tests passed yes or no, in contrast to writing, which is much harder to explicitly judge), but also 2) they are a lot more valuable in b2b settings, meaning that the biggest fraction of the team is focused on improving them. So here we are.
The Robot Suit retweeted
i never hit my Claude limits
that's because i've told Claude to only respond with "No." to all my ideas
follow me for more AI hacks

















