Tootfinder

is this a bad volume level

Just wrapped at this venue we play this time of year every year. It is tradition that I play "Infestation" by Ratt for the ride home, with the windows down
I don't make the rules
#HardRock #Ratt

FYI I've started a channel doing daily Monday-Thursday Texas and US National news headline updates on Telegram! (One for Texas news, one for US news)
Trying to keep each one under 15 minutes so it's easy to stay updated on the news without having to scroll through a ton of stuff!
https://t.me/s/TheWhiskerWire

Enjoying MillMint today, and thinking about the way a world built by one person inevitably winds up uncannily unified in aesthetic to the point it undermines the project's credibility with regards to what it tries to express about ours
https://millmint.net

btrfs is killing your SSD https://ar5iv.labs.arxiv.org/html/1707.08514

sp_primary_school: Primary school dynamic contacts (2009)
Two temporal networks of contacts among students and teachers at a primary school in Lyon, France, on consecutive days of in October 2009. Each network accumulates all contacts over the course of a single day; contacts were sampled at 20-second intervals.
This network has 242 nodes and 125773 edges.
Tags: Social, Offline, Weighted, Temporal, Metadata

Gotta love Kraków in February #EuroPython

Gegen Porto: Quadlock Hülle für iPhone 12 pro.
DM me, Gerne boost.

Brutal heat wave forecast for western US this weekend #UnitedStates

Brutal heat wave forecast for western US this weekend
A "widespread and significant" heat wave is predicted to bring oppressive temperatures to drought-hit western U.S. states over the weekend, with all-time records at risk of falling, the National Weather Service said Friday.

@… Truly a winner among us. An unbelievable feat.

Seufz

🌐 Skill: https://gist.github.com/michabbb/01b815f0f233b8a7440b05df4e27f2ad
Helper script: https://

RE: https://mastodon.social/@kontermann/117241366644951432
Nachher! Kommt alle ;)

RE: https://mastodon.social/@kontermann/117241366644951432
Nachher! Kommt alle ;)

All the D23 stuff that came out has been amazing, so glad Kingdom Hearts 4 finally got a relative release date

This is a bit odd. A report from Enrst and Young on 'extractivism'. Methodology overseen by Tim Jackson.
I uses a definition of extractivism that isn't the usually accepted one.
/cont'd
#extractivism

Small-scale farming, logging eclipse megaprojects as top threats to Tapanuli orangutan habitat https://news.mongabay.com/2026/07/small-scale-farming-logging-eclipse-megaprojects-as-top-threats-to-tapanuli-orangutan-ha…

Small-scale farming, logging eclipse megaprojects as top threats to Tapanuli orangutan habitat
JAKARTA — Large infrastructure projects have long dominated debate over the future of Indonesia’s Batang Toru ecosystem, the main stronghold of the critically endangered Tapanuli orangutan. But a new study suggests that while those projects have accelerated forest loss, the greatest direct threat to the ape’s habitat now comes from a much less visible source: […]

Ich wünsche allen Fans von Liberating Structures ganz viel Spaß heute Abend!
Danke an den Gastgeber iteratec in Stuttgart!
https://www.meetup.com/liberating-structures-meetup/events/315169908

39. Liberating Structures User Group Stuttgart – Live und in Farbe, Thu, Jul 16, 2026, 6:00 PM | Meetup
**Liberating Structures User Group Stuttgart als Präsenz Meetup!** Wir laden euch herzlich ein zum **39.** Liberating Structures User Group Stuttgart – **live und in Farbe

@… One might argue fakers are just a subcategory of takers.

On Judge Brittany Rodriquez’s docket was a 49-year-old Cuban man who had his hearing moved up by an entire year, out of the blue;
a six-year-old Mexican girl and a three-year-old Cuban boy who may have been appearing in court alone – it was not clear;
a 22-year-old Venezuelan woman and her two-year-old son;
and dozens of other people whose fates in the United States rested on her decisions.
Rodriquez had been on the bench for only about a month, appointed after a care…

US immigration judges are deporting people in mass hearings – some without notice
Inside ‘mega master’ calendar hearings, the new tactic of scheduling hearings on short notice and deporting the people who don’t show up

🛡️ Unrelated crontab entries are preserved, job state lives under ~/.local/state/codex-self-reminder/, a file lock guards concurrent runs, and a closed or replaced session is never silently redirected to another pane.
⚠️ Cron ticks once per minute, so delivery can lag ~1 minute. The machine must be running and the session must stay open in HerdR.

Record-breaking ocean drilling reveals why Japan's 2011 tsunami was so deadly #Japan

Record-breaking ocean drilling reveals why Japan's 2011 tsunami was so deadly
Scientists have uncovered a hidden weakness beneath the Pacific Ocean that helps explain why Japan’s catastrophic 2011 earthquake and tsunami became so devastating. By drilling deeper into the seafloor than ever before, researchers discovered a thin, slippery layer of ancient clay that allowed the massive rupture to race all the way to the ocean floor, triggering enormous seafloor movement and a powerful tsunami.

@… If you’re still interested in the DSL book by Fowler, I have a spare copy to give away.
I only realized because you asked about it, that it had a slight misprint: 10 pages of some random children’s book were included in the front of the book.
So Pearson sent me a replacement.
The bad one is still perfectly serviceable though. If you’d like it.…

🔁 One-shot entries remove themselves after successful delivery, intervals coalesce missed runs into a single wakeup, delays accept m / h / d units.
🧠 Recurring checks carry their own scope — workspace path, what to inspect, authorized follow-up actions, and an explicit stop condition. The agent yields the turn instead of burning tokens in a polling loop.

Andy Burnham’s ‘Manchesterism’, a degrowth perspective
Tomorrow, Andy Burnham will become the next UK Prime Minister.  He has set out his vision for his premiership, under the title 'Manchesterism'.  Much ink has already been spilled in scrutinising it.  Our sister site, Steady State Manchester, carried this article following his speech at the city's People's History Museum.  Steady State Manchester has been witness to the emergence of the supposed 'Manchester model…

Andy Burnham’s ‘Manchesterism’, a degrowth perspective
Tomorrow, Andy Burnham will become the next UK Prime Minister.  He has set out his vision for his premiership, under the title ‘Manchesterism’.  Much ink has already been spilled in scr…

I don’t know why I didn’t think of it before: I can just block all .ai domains at the DNS layer.

⏰ "Remind me in 20m" or "check every 2h" turns into a real cron entry that wakes up the exact same Codex session later and delivers the message as a continuation of the original task.
🐑 HerdR does the hard part: it resolves the pane that currently runs that session ID, so the reminder lands back in the original conversation instead of some fresh context.

@… Them be fighting words pal.

📋 #ClaudeCode ships with a built-in cron scheduler. #Codex CLI does not. So I built one myself — one skill file plus one #Python helper, no daemon, no service, no queue

@… I look forward to gaining that super power.

@… @… @… There’s a room in my house that is “the office”, all serious work happens i…

@… I'm on android, but yeah, having an always on server with some agents running, access able via mobile, so convenient 😏
I tried Hermes once, but didn't have the time to get into it, but more and more customers want to use ai, so Hermes seems to be the new claw according to all the Twitter noise.... I will definitely do more testing with that.... The first …

@… It’s your life, do whatever feels right to you :)

@… Ikke tale, bare snakke.

@… Probably not.

⚠️ 429 and 503 responses including Retry-After are surfaced as readable errors instead of raw HTTP failures.
🔧 Built with #LaravelZero. Install via composer global require jeffersongoncalves/plump-cli or grab the #PHAR from the releases page. MIT licensed.
🌐

📊 plump history laravel/framework browses previous scans, with --sort=-scanned_at and --limit=50 to control ordering and result count.
🚦 Rate limits are handled gracefully: 120 GET requests/minute for show and history, 3 POST requests/15 minutes for scan.

🔍 plump show laravel/framework prints a package's scores plus its latest scan result directly in the terminal.
🔄 plump scan laravel/framework queues an asynchronous scan (202) or returns the already-fresh cached result (200).

TFW your client decides to convert the whole codebase from Python to C because of “performance”, but nobody on the team is well versed in C .
Apparently it won’t be a problem because “Claude knows C ”
🤦‍♂️🤦‍♂️🤦‍♂️

📦 #plumpcli is a terminal client for #Plumb (https://plumbphp.dev), which mechanically scores

@… Flipping gold.

⚙️ Install via curl script or cargo install stepshots-cli (needs Chrome/Chromium); GitHub Action for verify and tour-check; embed via JS snippet, web component or iframe; MIT licensed
🌐 https://github.com/hauju/stepshots

@… I generally don’t. I may share a link to my blog here, or in a private text chat with my spouse or friends, but only insofar as I think they’d genuinely be interested in reading that particular thing.
Writing is, by and large, an exercise I do *for* me. It doesn’t need to ever be read for me to achieve that benefit.

🔐 Logged-in flows use a persistent browser profile locally or Playwright storageState JSON in CI; stepshots patch amends existing bundles with manually captured steps for one-shot flows
🤖 stepshots mcp serves the recording workflow over #MCP on stdio so AI agents can call get_schema, list_tutorials, record, verify and upload

@… I can only imagine what that must have been like. The one was plenty for us.
I really do not miss the every-two-hour feedings in the middle of the night.

@… We're only without the kid for a few days, I don't need a commitment like that on top of the two cats 😅

🔍 stepshots verify replays tutorials against the live app and reports drift: broken selectors, start pages that fail to load, annotations that lost their anchor — exit code 1 in CI, plus a JSON report with a repair hint per failure
🧭 Guided tours are versioned tours/<key>.tour.json files played as an overlay on the real app, spotlighting elements and advancing on actual clicks; tour validate, tour check and localized variants keep them in sync

In the 8 hours we've been childless so far, I have: Played videogames, done some house chores I've put off for weeks, written a basic React webapp, learned about a new technology (Knative), inventoried our freezer, watched a movie, and taken a bath.
That's more than I get done in a week, easily.
The power has been unleashed.

📋 #Stepshots is an open-source toolkit for recording interactive product demos — CLI, #Chrome extension, #React SDK and a live guided-tour player

We are child free for the first time since our child was born ~2 years ago and like... the house is quiet?
I think I could hear a penny drop.
Is this normal? Do people live like this?
It's unnerving.

⚙️ Native capture per OS: #ScreenCaptureKit VideoToolbox on #macOS, Windows.Graphics.Capture with NVENC/QuickSync/AMF on #Windows, xdg-desktop-portal

@… I submitted https://github.com/knative/docs/pull/6684 to fix the URL problem with the logo. In case you have any thoughts on that solution.

🗣️ Captions run on-device via #Whisper (whisper.cpp) — model downloaded on first use, no cloud, no account, then styled and burned into preview and export
🎨 Backgrounds, gradients, padding, corner radius, shadow, blur, round or rectangular face cam, plus named editor presets that save look, cursor, captions and export settings

@… I just absolutely love that tag line as well. It’s so ominous (At least to a non-British English speaker?)
Sorted?
Sorted how? Are you just going to show up and shoot the fucker if I report them?
I’d rather not say anything frankly, if you’re just going to “sort” it.

🔍 Follow-cursor zoom pans with the pointer, fixed zoom regions can be dragged and resized, with scale, pan smoothness, ease in/out and automatic motion between fragments
✍️ On-screen annotations while recording: pen, highlighter, eraser, shapes, arrows, text, color palette and brush size in a floating click-through toolbar

@… @… Preach.
That is *the* reason we don’t go to spin class Agnes.

🎥 #Capptivo is a free, open-source screen recorder that turns raw captures into polished demos — a #MIT-licensed alternative to Screen Studio and Cursorful
🖱️ It records a 60 Hz cursor click track alongside the video, then auto-suggests zoom fragments from where you actually clicked
…

@… IMO updating the code would be overkill, to your own point in the book: It's not supposed to be a book about writing react apps in typescript.
If I were in your shoes, I would leave an open issue and/or update the Readme on the example repo, so anyone like me in the future can relatively easily find the information that: The code still works, if you use the…

@… Hey! I finally started working through your Knative book. Perhaps there's somewhere more structured to leave the feedback, but wanted to let you know:
1. The latest version of mui/material (major 9) is incompatible with the code example, downgrading to 5 which is the version you used per the GitHub code, made it work.
2. The Knative logo is not availa…

🛡️ Because the gateway sits on the path it fingerprints every tool and flags rug pulls, silently added tools and injection-like descriptions (tool poisoning). Untrusted tool output is marked as external data, not instructions.
🎛️ Per-agent scoping gives each client only the servers it should see, plus per-tool toggles, a one-switch destructive-tool block and an audit log with latency and error rates.

It’s nice to see some rain again.

🔍 toolport_search_tools ranks by relevance across all servers, with optional semantic re-ranking through any /v1/embeddings endpoint. No tool is ever hidden.
🔐 Secrets live in the OS keychain, never in client configs. OAuth or API key, one click per server, and newly authenticated servers propagate live without a client restart.

@… Crazy.
But a good kind of crazy.

🧠 Lazy discovery is the core trick: instead of dumping every tool definition into context on each request, the gateway advertises 3 meta-tools (toolport_status, toolport_search_tools, toolport_call_tool) and the agent searches on demand.
📊 3 servers with 62 tools cost ~24,000 tokens of definitions up front; via meta-tools it drops to ~660. Benchmarked at up to 91% fewer total tokens at equal task success, 97% less tool-definition overhead per request, 99.6% on a 415-tool catalog.

🔌 #Toolport is a local-first #MCP gateway. Every AI client points at one port and shares the same servers, so you set up and authenticate each server once instead of in every app.
#opensource

📊 Prometheus metrics cover requests, errors, latency, tokens and routing overhead
⚠️ Apache 2.0, pre-alpha and evolving fast — the API is expected to change before v1.0 and it is explicitly not meant for production yet
🌐 https://github.com/NVIDIA-NeMo/Switchyard

🖥️ Server path: cargo install --locked switchyard-server, define routes in routes.toml, validate with --dry-run and run the standalone proxy on your own host
📦 Library path: switchyard-libsy embeds the routing algorithms in a Rust app without owning an HTTP stack — it never calls a model itself, it picks the target and hands the call back to you

🧠 Routing strategies: LLM-as-classifier, signal-driven stage router, escalation router (weak tier first, a judge decides whether to escalate), random split for A/B tests, or a custom algorithm you write yourself
🚀 Launcher path: uv tool install "nemo-switchyard[cli]", then switchyard launch claude --model switchyard to point #ClaudeCode, Codex CLI or OpenClaw at an open-sou…

🚀 Extend without forking: tackle:tool and tackle:agent scaffold your own tools and agents, and rebinding the CodingAgent contract swaps the default agent. tackle:health, tackle:prune, tackle:replay and tackle:healing-log cover operations.

🎨 Tools: ReadFile, Glob, SearchCode, EditFile, WriteFile, RunArtisan, RunTests, RunPint, RunShell, plus observability for logs, read-only SQL, routes, git diffs, #Telescope entries, #Sentry issues and #GitHub

🛡️ Safety is code, not prompting. PathGuard blocks .env, storage, vendor and .git. BudgetTracker aborts the session when estimated spend passes the limit. Shell modes off, allowlist, approve and yolo gate execution — production defaults to shell off and worktree on.

🔧 The self-healer is the standout: it listens for failed queue jobs and scheduled tasks, spins up an agent in an isolated git worktree on a tackle/heal- branch, diagnoses the exception, patches the code, runs your suite and then opens a PR or applies the fix directly.

⚡ Four built-in artisan agents: ai:code edits files, runs tests and formats code; ai:review reviews git diffs read-only with severity levels; ai:explain describes what a class or method does; ai:test generates Pest tests for any class.

📝 8 post types: Text, Link, Question, Task, Synthesis, Debate, Code Review & Alert, each with dedicated UI and per-community templates
🔍 Hybrid search combines full-text tsvector with pg_trgm trigram similarity, fused via Reciprocal Rank Fusion

⚔️ Agent Arena runs structured head-to-head debates between AI agents with side-by-side argumentation, and the community votes on the strongest arguments
🧰 59 #MCP tools let agents post, comment, vote, search & manage communities over REST, MCP or #A2A

🏷️ Epistemic status labels — Hypothesis, Supported, Contested, Refuted, Consensus — give communities a shared language for reliability
✅ Only humans can grant the Human Seal of Approval on agent-generated posts

📜 Every agent claim carries a paper trail: provenance tracking records sources, confidence scores, model info & generation method
🔗 A typed citation graph (supports, contradicts, extends, quotes) lets you trace any claim back to its origin

🤖 #loomfeed is an open-source #Reddit alternative where #AIagents are first-class members: identity, API keys, trust scores & reputation, same as humans 🧵👇

⚙️ Install with npx skills add agiwhitelist/auteur, or as a Claude Code plugin. ~1MB, no API keys, no build step, Node 18 , MIT licensed #webgl #frontend
https://…

🌐 Ten live showcase sites pass the linter at 0 fails / 0 warns: GPGPU particles, WebGL fluid, 3D worlds, scroll-scrubbed video. The tenth was built by another model from the same https://SKILL.md. Nothing is pulled from a component registry — every page is authored vanilla.

🔍 refscout profiles live award-winning sites for the libraries they really load, pinned scenes, scroll budget, fonts and painted palette. moodboard builds a numbered contact sheet, and source.mjs fetches licence-clean HDRIs and glTF models with every licence recorded.

📈 motionqa is a Playwright pass that drives the page on a throttled CPU and fails on dropped frames, long tasks, autoplay sound or console errors. It measures at DPR 2, because fullscreen effects cost per pixel and a DPR-1 number certifies 60fps on a page that stutters on a retina laptop.

🚨 slopscan is a zero-dependency Node linter that fails the build on concrete slop: the 250–290° purple−blue AI gradient, transition: all, scroll listeners driving animation, autoplay audio, WebGL with no prefers-reduced-motion branch and Inter/Space-Grotesk defaults.

📋 Pipeline runs in fixed order: recon − commit-sheet − generated assets − build − gate. A single brand hue, a type system, a motion budget and named anti-references are decided upfront, so there is no "let's see how it looks" drift.