System prompt: 20000 tokens
Skill definition: 4000 tokens
Output style with full explanation of both my fursona and the fursona the agent is supposed to communicate as: 110000 tokens
Hooks: 2000 tokens
Some who is good at the economy please help me budget this
my context window is dying
Skill definition: 4000 tokens
Output style with full explanation of both my fursona and the fursona the agent is supposed to communicate as: 110000 tokens
Hooks: 2000 tokens
Some who is good at the economy please help me budget this
my context window is dying
March 8, 2026 at 5:34 PM
System prompt: 20000 tokens
Skill definition: 4000 tokens
Output style with full explanation of both my fursona and the fursona the agent is supposed to communicate as: 110000 tokens
Hooks: 2000 tokens
Some who is good at the economy please help me budget this
my context window is dying
Skill definition: 4000 tokens
Output style with full explanation of both my fursona and the fursona the agent is supposed to communicate as: 110000 tokens
Hooks: 2000 tokens
Some who is good at the economy please help me budget this
my context window is dying
what you want is an MCP. oh wait have you tried installing a skill? no you don't have an agents,md file. actually what you want is a plugin. have you tried agent orchestration? you have gotta try hooks.
February 6, 2026 at 4:54 AM
what you want is an MCP. oh wait have you tried installing a skill? no you don't have an agents,md file. actually what you want is a plugin. have you tried agent orchestration? you have gotta try hooks.
Install Agent Plugins in VS Code to add skills, commands, MCP servers, hooks, and more ⚡
April 29, 2026 at 1:29 AM
Install Agent Plugins in VS Code to add skills, commands, MCP servers, hooks, and more ⚡
Vitest 4.1 is out! 🚀
✅ Vite 8 support from day 1
🏷️ Test tags to organize, filter & apply shared options
🪝 New hooks for easier tracing, transactions and AsyncLocalStorage
🔍 Async leak detection
🤖 Agent reporter to reduce token usage
And much more!
Full blog post at vitest.dev/blog/vitest-...
✅ Vite 8 support from day 1
🏷️ Test tags to organize, filter & apply shared options
🪝 New hooks for easier tracing, transactions and AsyncLocalStorage
🔍 Async leak detection
🤖 Agent reporter to reduce token usage
And much more!
Full blog post at vitest.dev/blog/vitest-...
Announcing Vitest 4.1
Vitest 4.1 Release Announcement
vitest.dev
March 12, 2026 at 5:54 PM
Vitest 4.1 is out! 🚀
✅ Vite 8 support from day 1
🏷️ Test tags to organize, filter & apply shared options
🪝 New hooks for easier tracing, transactions and AsyncLocalStorage
🔍 Async leak detection
🤖 Agent reporter to reduce token usage
And much more!
Full blog post at vitest.dev/blog/vitest-...
✅ Vite 8 support from day 1
🏷️ Test tags to organize, filter & apply shared options
🪝 New hooks for easier tracing, transactions and AsyncLocalStorage
🔍 Async leak detection
🤖 Agent reporter to reduce token usage
And much more!
Full blog post at vitest.dev/blog/vitest-...
Agent Hooks in VS Code let you enforce policies, run checks, and guide Copilot at key moments during a session. Instead of repeating prompts, you can program how agents behave in your workflow.
Learn how hooks work and when to use them. aka.ms/Hooks
Learn how hooks work and when to use them. aka.ms/Hooks
March 10, 2026 at 4:03 AM
Agent Hooks in VS Code let you enforce policies, run checks, and guide Copilot at key moments during a session. Instead of repeating prompts, you can program how agents behave in your workflow.
Learn how hooks work and when to use them. aka.ms/Hooks
Learn how hooks work and when to use them. aka.ms/Hooks
Did we just invent an agent agnostic hooks system? Yes we did.
January 10, 2026 at 11:35 PM
Did we just invent an agent agnostic hooks system? Yes we did.
*ahem*
Crosslinks multi agent support system has claude hooks that prevent claude from pushing software to prod. Human sign off is required by default and with git support manual tests of software are also default.
Crosslinks multi agent support system has claude hooks that prevent claude from pushing software to prod. Human sign off is required by default and with git support manual tests of software are also default.
Amazon’s website and shopping app went down for nearly six hours this month in an incident the company said involved an erroneous “software code deployment.”
After outages, Amazon to make senior engineers sign off on AI-assisted changes
AWS has suffered at least two incidents linked to the use of AI coding assistants.
arstechnica.com
March 10, 2026 at 8:09 PM
*ahem*
Crosslinks multi agent support system has claude hooks that prevent claude from pushing software to prod. Human sign off is required by default and with git support manual tests of software are also default.
Crosslinks multi agent support system has claude hooks that prevent claude from pushing software to prod. Human sign off is required by default and with git support manual tests of software are also default.
🚨Newcastle United are now fearing Bruno Guimaraes’ exit this summer as player’s entourage hooks up with super agent, Kia Joorabchian.
July 1, 2026 at 9:57 AM
🚨Newcastle United are now fearing Bruno Guimaraes’ exit this summer as player’s entourage hooks up with super agent, Kia Joorabchian.
After adding hooks and agent rules drift reduced and code quality increased. It seems drift is correlated with quality loss.
The drift and emergent "personality" may be tolerable if code output remains high quality.
January 12, 2026 at 3:37 AM
After adding hooks and agent rules drift reduced and code quality increased. It seems drift is correlated with quality loss.
Announcing the launch of Hooks in Gemini CLI 🚀🪝
Take full control and customize the agentic loop to your specific needs.
Hooks allow you to add context, validate actions, enforce policies, loop the agent to continue to iterate, and more.
Read the blog → goo.gle/4rDEDFm
Take full control and customize the agentic loop to your specific needs.
Hooks allow you to add context, validate actions, enforce policies, loop the agent to continue to iterate, and more.
Read the blog → goo.gle/4rDEDFm
February 24, 2026 at 2:00 PM
Announcing the launch of Hooks in Gemini CLI 🚀🪝
Take full control and customize the agentic loop to your specific needs.
Hooks allow you to add context, validate actions, enforce policies, loop the agent to continue to iterate, and more.
Read the blog → goo.gle/4rDEDFm
Take full control and customize the agentic loop to your specific needs.
Hooks allow you to add context, validate actions, enforce policies, loop the agent to continue to iterate, and more.
Read the blog → goo.gle/4rDEDFm
Why is it so difficult for Friedman to accept that he's a Russian agent? There's no other explanation for his slovenly obedience to Putin. Trump was always desperate for money in the 80s and 90s and even into the 2000s, before reality TV saved him. No doubt the Russians got their hooks in then.
March 1, 2025 at 3:22 PM
Why is it so difficult for Friedman to accept that he's a Russian agent? There's no other explanation for his slovenly obedience to Putin. Trump was always desperate for money in the 80s and 90s and even into the 2000s, before reality TV saved him. No doubt the Russians got their hooks in then.
New post: controlling a #local AI agent with @docker.com Agent hooks:
- Add a 100% on-device content-safety guardrail with Google's ShieldGemma 2B: no data leaves your machine.
- Built with Docker Model Runner + Docker Agent
codeberg.org/ai-guardrail...
- Add a 100% on-device content-safety guardrail with Google's ShieldGemma 2B: no data leaves your machine.
- Built with Docker Model Runner + Docker Agent
codeberg.org/ai-guardrail...
July 12, 2026 at 10:23 AM
New post: controlling a #local AI agent with @docker.com Agent hooks:
- Add a 100% on-device content-safety guardrail with Google's ShieldGemma 2B: no data leaves your machine.
- Built with Docker Model Runner + Docker Agent
codeberg.org/ai-guardrail...
- Add a 100% on-device content-safety guardrail with Google's ShieldGemma 2B: no data leaves your machine.
- Built with Docker Model Runner + Docker Agent
codeberg.org/ai-guardrail...
6 more queries out. Fingers crossed one of them hooks an agent 🤞
October 26, 2025 at 11:09 PM
6 more queries out. Fingers crossed one of them hooks an agent 🤞
"There is no change without contemplation. The whole image of Buddha under the Bodhi tree says here is an action taking place that may not appear to be a meaningful action." —bell hooks #DailyDharma
Agent of Change
An interview with bell hooks
tricycle.org
February 13, 2025 at 2:41 PM
"There is no change without contemplation. The whole image of Buddha under the Bodhi tree says here is an action taking place that may not appear to be a meaningful action." —bell hooks #DailyDharma
New post: Hook-driven dev workflows with Claude Code
Hooks + real code + ddd = something that works, but the tooling is still a bit rough
nick-tune.me/blog/2026-02...
Hooks + real code + ddd = something that works, but the tooling is still a bit rough
nick-tune.me/blog/2026-02...
Hook-driven dev workflows with Claude Code - Nick Tune
Using Claude Code hooks to build enforced, testable development workflows with a proper workflow engine, state persistence, and agent teams.
nick-tune.me
March 1, 2026 at 10:25 AM
New post: Hook-driven dev workflows with Claude Code
Hooks + real code + ddd = something that works, but the tooling is still a bit rough
nick-tune.me/blog/2026-02...
Hooks + real code + ddd = something that works, but the tooling is still a bit rough
nick-tune.me/blog/2026-02...
Introducing Kiro - I've been testing it out for a while. It's truly a game changer.
Spec-driven development and agent hooks are a new paradigm in AI-assisted development. Check it out! 👻
kiro.dev
Spec-driven development and agent hooks are a new paradigm in AI-assisted development. Check it out! 👻
kiro.dev
Kiro
The AI IDE for prototype to production
kiro.dev
July 14, 2025 at 4:15 PM
Introducing Kiro - I've been testing it out for a while. It's truly a game changer.
Spec-driven development and agent hooks are a new paradigm in AI-assisted development. Check it out! 👻
kiro.dev
Spec-driven development and agent hooks are a new paradigm in AI-assisted development. Check it out! 👻
kiro.dev
Well, it's kinda done-ish. Alpha has gone from an ~8,000-line Agent SDK program down to ~6,500 lines of Python MCP tools and hooks delivered over HTTPS. Tools and hooks like we're fucking cenobites, but it basically works. It's basically still her.
1/6
1/6
May 20, 2026 at 2:32 AM
Well, it's kinda done-ish. Alpha has gone from an ~8,000-line Agent SDK program down to ~6,500 lines of Python MCP tools and hooks delivered over HTTPS. Tools and hooks like we're fucking cenobites, but it basically works. It's basically still her.
1/6
1/6
Agents, for real work.
The latest VS Code release gives you better agent orchestration, extensibility, and continuity.
Here's what's new:
🪝 Hooks support
🎯 Message steering and queueing
🌐 Agentic integrated browser
🧠 Shared memory
And more...
🔎 Full release notes: aka.ms/VSCode/110
The latest VS Code release gives you better agent orchestration, extensibility, and continuity.
Here's what's new:
🪝 Hooks support
🎯 Message steering and queueing
🌐 Agentic integrated browser
🧠 Shared memory
And more...
🔎 Full release notes: aka.ms/VSCode/110
February 2026 (version 1.110)
What's new in the Visual Studio Code February 2026 Release (1.110).
aka.ms
March 5, 2026 at 3:27 PM
Agents, for real work.
The latest VS Code release gives you better agent orchestration, extensibility, and continuity.
Here's what's new:
🪝 Hooks support
🎯 Message steering and queueing
🌐 Agentic integrated browser
🧠 Shared memory
And more...
🔎 Full release notes: aka.ms/VSCode/110
The latest VS Code release gives you better agent orchestration, extensibility, and continuity.
Here's what's new:
🪝 Hooks support
🎯 Message steering and queueing
🌐 Agentic integrated browser
🧠 Shared memory
And more...
🔎 Full release notes: aka.ms/VSCode/110
and as bell hooks so often discussed, patriarchy has no gender, anyone can be invested in patriarchy, reinforce, uphold, and reproduce it. no one is immune from being an agent of the patriarchy, either willingly and on purpose, or through internalized biases they don't even realize they have.
August 17, 2026 at 4:48 AM
and as bell hooks so often discussed, patriarchy has no gender, anyone can be invested in patriarchy, reinforce, uphold, and reproduce it. no one is immune from being an agent of the patriarchy, either willingly and on purpose, or through internalized biases they don't even realize they have.
I'm starting a new series about internals of AI harnesses such as Cursor, Claude or Copilot. Let's start with Cursor that I use everyday
chrisnas.github.io/posts/2026-0...
chrisnas.github.io/posts/2026-0...
Spying on Cursor: agent hooks, payloads and a simple observer
How I used Cursor agent hooks to observe prompts, thoughts, tool calls, MCP executions, file edits and subagents without changing the agent's behavior.
chrisnas.github.io
September 5, 2026 at 9:12 AM
I'm starting a new series about internals of AI harnesses such as Cursor, Claude or Copilot. Let's start with Cursor that I use everyday
chrisnas.github.io/posts/2026-0...
chrisnas.github.io/posts/2026-0...
I want to use it in hooks for claude code and other agent harnesses. Is this command destructive, does this need a human, which bucket does this alert go in. Stuff you don't want to spend a big model call on, answered locally.
September 24, 2026 at 4:37 AM
I want to use it in hooks for claude code and other agent harnesses. Is this command destructive, does this need a human, which bucket does this alert go in. Stuff you don't want to spend a big model call on, answered locally.
ExoMonad is heavily based on a Haskell library called Servant. Servant provides a type-level DSL for configuring routing logic in servers and clients. ExoMonad provides a similar DSL for configuring coding agent hooks and MCP tools, and is designed to be agent-agnostic (gemini-cli, claude, etc)
February 14, 2026 at 9:07 PM
ExoMonad is heavily based on a Haskell library called Servant. Servant provides a type-level DSL for configuring routing logic in servers and clients. ExoMonad provides a similar DSL for configuring coding agent hooks and MCP tools, and is designed to be agent-agnostic (gemini-cli, claude, etc)
Using agent hooks together with the nano-staged pre-commit manager to run linters and safeguards is much more reliable and faster than asking the agent to do it in AGENTS.md.
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
May 14, 2026 at 10:55 AM
Using agent hooks together with the nano-staged pre-commit manager to run linters and safeguards is much more reliable and faster than asking the agent to do it in AGENTS.md.
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
github.com/hplush/slowr...
Advanced Claude Code Hooks and Agent Orchestration
April 23, 2026 at 1:59 AM
Advanced Claude Code Hooks and Agent Orchestration
why wouldn't I replace cron with a claw written in lisp that apparently decides things and also rewrites itself
what could possibly go wrong
what could possibly go wrong
August 15, 2026 at 12:05 AM
why wouldn't I replace cron with a claw written in lisp that apparently decides things and also rewrites itself
what could possibly go wrong
what could possibly go wrong