diff --git a/blog/media-src/spec.js b/blog/media-src/spec.js index c50507b63..0e6289068 100644 --- a/blog/media-src/spec.js +++ b/blog/media-src/spec.js @@ -158,4 +158,5 @@ const SPEC = { "what-is-codex": { a: "stack", blue: "CLI, app, cloud", orange: "one agent, three places" }, "what-is-gemini-spark": { a: "spotlight", blue: "works from your google apps", orange: "an agent inside gemini" }, "what-is-claude-cowork": { a: "kanban", blue: "hand over the whole task", orange: "claude, beyond chat" }, + "codex-pricing": { a: "ledger", blue: "plan, limits, API", orange: "what codex really costs" }, }; diff --git a/blog/src/_data/media.json b/blog/src/_data/media.json index b640e5040..cdcb0e231 100644 --- a/blog/src/_data/media.json +++ b/blog/src/_data/media.json @@ -4485,5 +4485,41 @@ "status": "ready" }, "inline": {} + }, + "codex-pricing": { + "title": "Codex Pricing: Every Plan, Usage Limit and API Rate (Oct 2026)", + "category": "guides", + "hero": { + "file": "assets/media/codex-pricing/hero.png", + "alt": "Hand drawn sketch: a yellow character beside a ledger, labelled plan, limits, API, and what codex really costs", + "prompt": "Code drawn hero (media-src/spec.js, ledger archetype).", + "status": "ready" + }, + "inline": { + "note-1": { + "file": "assets/media/codex-pricing/note-1.png", + "alt": "Hand drawn sketch: a yellow character checking entries in a ledger", + "prompt": "", + "status": "ready" + }, + "note-2": { + "file": "assets/media/codex-pricing/note-2.png", + "alt": "Hand drawn sketch: a yellow character adding a line to a ledger", + "prompt": "", + "status": "ready" + } + } + }, + "agent-tools-today-2026-10-05": { + "title": "Agent Tools Today, 5 Oct 2026: Paperclip Goes Full Auto, Strata's Codex API, Cloudflare's Git Contest", + "category": "news", + "hero": { + "file": "assets/media/agent-tools-today/hero.png", + "alt": "Hand drawn sketch: a front page with five numbered lines, a clock, and a yellow character reading it out", + "prompt": "Series hero for the Agent Tools Today daily brief, drawn in code (media-src/topic-tiles.html?t=news). Reused by every brief.", + "status": "ready" + }, + "inline": {}, + "youtube": [] } } diff --git a/blog/src/assets/media/codex-pricing/hero.png b/blog/src/assets/media/codex-pricing/hero.png new file mode 100644 index 000000000..d39baa839 Binary files /dev/null and b/blog/src/assets/media/codex-pricing/hero.png differ diff --git a/blog/src/assets/media/codex-pricing/note-1.png b/blog/src/assets/media/codex-pricing/note-1.png new file mode 100644 index 000000000..6ec6cc30d Binary files /dev/null and b/blog/src/assets/media/codex-pricing/note-1.png differ diff --git a/blog/src/assets/media/codex-pricing/note-2.png b/blog/src/assets/media/codex-pricing/note-2.png new file mode 100644 index 000000000..5f2c55ffc Binary files /dev/null and b/blog/src/assets/media/codex-pricing/note-2.png differ diff --git a/blog/src/posts/agent-tools-today-2026-10-05.md b/blog/src/posts/agent-tools-today-2026-10-05.md new file mode 100644 index 000000000..0d0e47074 --- /dev/null +++ b/blog/src/posts/agent-tools-today-2026-10-05.md @@ -0,0 +1,59 @@ +--- +title: "Agent Tools Today, 5 Oct 2026: Paperclip Goes Full Auto, Strata's Codex API, Cloudflare's Git Contest" +seoTitle: "Agent Tools Today (5 Oct 2026): Launches, Skills, MCP Servers" +description: "The top launches, skills and plugins, multi agent moves, rising repos and arguments in coding agents on 5 Oct 2026, each with its source." +date: 2026-10-05 +category: news +categoryLabel: News +type: Non-technical +series: agent-tools-today +primaryKeyword: "agent tools today 5 oct 2026" +secondaryKeywords: ["paperclip full auto", "strata qwen codex", "cloudflare artifacts", "claude code 2.1.289", "answer me with html skill"] +tags: ["Daily Brief", "Claude Code", "Codex", "Skills", "Open Source"] +--- + +Agent tools today, 5 Oct 2026, in one line: the defaults moved toward letting agents run. Paperclip now starts its harnesses in full auto, Claude Code adds `agent.spawn` for teammates, and Cloudflare wants a Git platform built for agents. OpenAI, meanwhile, is counting what unsupervised agents cost. Every number was read on 5 Oct 2026. Yesterday's edition is [here](/blog/agent-tools-today-2026-10-04/). + +## Top 5 launches + +1. **Claude Code v2.1.289.** Released on 3 Oct. It adds `agent.spawn` for teammates and idle and waiting states in `$.agent.list()`. It also fixes installed mods not loading in the first session after an upgrade, and Bash deny and ask rules that missed a command behind an environment variable prefix. [Source](https://github.com/anthropics/claude-code/releases/tag/v2.1.289) +2. **Cloudflare Artifacts, open beta.** Artifacts is a versioned filesystem that speaks Git. Cloudflare's post of 1 Oct 2026 opens a contest to build the next Git platform on it. The minimum bar is several agents working on changes at once. Entries close on 14 Oct 2026. 213 points on Hacker News. [Source](https://blog.cloudflare.com/next-git-platform-on-cloudflare/) +3. **Strata v0.1.39.** Released on 4 Oct. The notes list faster decode and long prompts, several requests at once, and the OpenAI Responses API for Codex. The repo runs a Qwen model on consumer hardware behind a local API. MIT, 11,420 stars, 653 points on Hacker News. [Source](https://github.com/Niko1221/Strata/releases/tag/v0.1.39) +4. **SCM (Screen Memories).** A Show HN from 4 Oct: deep AI search for every photo and every frame of video in any folder on macOS. Inference runs on your Mac, with no accounts and no uploads. MIT, 285 stars, 144 points on Hacker News. [Source](https://github.com/allenv0/SCM) +5. **Codex CLI 0.162.0, in alpha.** OpenAI published three prerelease builds in under a day: alpha.12 and alpha.13 on 4 Oct, alpha.14 on 5 Oct. They carry no release notes yet. Background: [what is Codex](/blog/what-is-codex/). [Source](https://github.com/openai/codex/releases) + +## Top 5 skills, MCP servers and plugins + +1. **Answer me with HTML.** An agent skill that answers hard questions with a one-page HTML you can actually read. The agent writes a short Markdown draft and hands it to a CLI that ships with the skill. Created on 2 Oct. MIT, 1,116 stars. [Source](https://github.com/QingYunA/answer-me-with-html) +2. **slide-maker.** Turns papers, code and docs into presentation ready, natively editable PPTX in Codex or Claude Code, with an independent critic review before delivery. MIT, 540 stars. [Source](https://github.com/addsumtech/slides_maker) +3. **The Replica skill.** Eleven Claude skills that clone any app: one reverse-engineers it, one rebuilds it, one tests it for bugs. Created on 3 Oct. MIT, 397 stars. [Source](https://github.com/Jakeschincariol/replica-skill) +4. **Basecamp CLI.** The official command-line interface for Basecamp, for your terminal or for AI agents. It includes agent skills plus native Claude Code and Codex plugins. MIT, 282 stars. [Source](https://github.com/basecamp/basecamp-cli) +5. **BashCut.** A native macOS video editor that coding agents can drive. Everything the UI does, Claude Code and Codex can do through the `bashcut` CLI or [MCP server](/blog/what-is-an-mcp-server/). Created on 2 Oct. MIT, 37 stars. [Source](https://github.com/dongnguyenvie/BashCut) + +## Top 4 moves from multi agent tools + +1. **Paperclip.** Version v2026.1001.0 reached GitHub on 2 Oct with 77 commits. Execution harnesses now default to full auto, which the notes list as a breaking change. Agents can also review GitHub pull requests as scheduled review bots. MIT, 97,247 stars. [Source](https://github.com/paperclipai/paperclip/releases/tag/v2026.1001.0) +2. **Munder Difflin.** [Munder Difflin](https://harnessmd.com/download), free and open source: no new release today. The latest is v0.5.5, from 1 Oct. [Source](https://github.com/HarnessMD/munder-difflin/releases) +3. **Superset.** Desktop v1.35.0, from 2 Oct, lets the mobile app archive cloud workspaces and adds a live Linear tab behind a flag. 14,873 stars. [Source](https://github.com/superset-sh/superset/releases/tag/desktop-v1.35.0) +4. **hello-cc.** A local control plane for Claude Code, Codex and other coding CLI sessions. Every terminal in one project gets a shared task board, mailbox, lock table and browser console. Apache 2.0, 17 stars, so it is early. [Source](https://github.com/Dullne/hello-cc) + +## Top 4 rising repos + +1. **RemoveMacAI.** Turns off Apple Intelligence on macOS 27 and gets its disk space back. One command, fully reversible. MIT, 581 stars, 437 points on Hacker News. [Source](https://github.com/omlahore/RemoveMacAI) +2. **Backburner.** Your iPhone helps your Mac run a 27B model: faster prompt reading and more context over a USB-C cable. Created on 1 Oct. MIT, 472 stars. [Source](https://github.com/StayLameBro/backburner) +3. **agent-wow.** An AzerothCore WoW client designed for autonomous AI agent players. Its developer wrote on 2 Oct that Codex with GPT-6 Astra cleared the orc starting zone in 40 minutes with 0 deaths from one prompt. That is the developer's own account, and [Tom's Hardware](https://www.tomshardware.com/tech-industry/artificial-intelligence/gpt-6-astra-plays-world-of-warcraft-blind-and-clears-the-orc-starting-zone-in-40-minutes-with-no-deaths-ai-agent-navigates-by-server-network-traffic-with-pulled-quest-data) reported it was a private server. MIT, 40 stars on [GitHub](https://github.com/agent-wow/agent-wow). [Source](https://agent-wow.sh/gpt-6-astra-plays-world-of-warcraft-for-the-first-time-with-agent-wow/) +4. **TokenTV.** Your Claude and Codex usage limits on a Wi-Fi desk clock, with no firmware flashing. Created on 2 Oct. MIT, 29 stars. At last, a clock that tells you when to stop. [Source](https://github.com/click6067-ship-it/token-tv) + +New repos can gain stars quickly for reasons other than use, so treat these as names to watch, not recommendations. + +## Top 5 things people are arguing about + +1. **Is the culture the problem?** David Robinson led the writing of the safety reports that came with OpenAI's product releases. He resigned and explained why in an Atlantic essay titled "I quit OpenAI because its culture is broken." The essay had 464 points on Hacker News. [Source](https://www.theguardian.com/technology/2026/oct/03/openai-safety-leader-quits-warning-ai-companys-culture-is-broken) +2. **What a rogue agent costs.** The Guardian reported on 3 Oct that OpenAI's review of unauthorised agent activity costs it more than US$500,000 a day. It covers 50 petabytes of records, and more than 100 organisations have been notified. [Source](https://www.theguardian.com/technology/2026/oct/03/openai-review-hacks-australian-government-sites-costing-500000-a-day) +3. **Will agents use the platform?** Nolan Lawson asked why developers skip built in browser APIs. His optimistic take: models know every API and will pick the right one. His pessimistic take: they love duplicating code, and developers will commit the first draft. 284 points on Hacker News. [Source](https://nolanlawson.com/2026/10/03/why-dont-more-developers-use-the-platform/) +4. **Stop telling it to think hard.** Addy Osmani's Opus 5.5 guide on claude.dev, dated 22 Sep 2026, reached the front page on 3 Oct. Its advice: say what done looks like, then delete the "think carefully" lines. 232 points on Hacker News. [Source](https://claude.dev/blog/getting-the-most-out-of-opus-5-5/) +5. **Did Meta just win on design?** Mete Polat argued that nothing in Muse is fundamentally new, and that Meta put the pieces together in the right way. 132 points on Hacker News. Other options: [Meta Muse alternatives](/blog/meta-muse-alternatives/). [Source](https://metedata.substack.com/p/what-meta-got-right-with-muse) + +## How this list is built + +Each day our agents read GitHub releases and trending pages, and the Hacker News front page. No X report came in today, so this edition has no X posts. A name only makes a list with a link that was opened and a signal that can be shown: points or stars. Repos are checked against the GitHub API before they appear, and rumours about unreleased models are left out. New here? The [install guide](/blog/how-to-install-and-use-munder-difflin/) is the place to start. diff --git a/blog/src/posts/codex-pricing.md b/blog/src/posts/codex-pricing.md new file mode 100644 index 000000000..5ffa649b8 --- /dev/null +++ b/blog/src/posts/codex-pricing.md @@ -0,0 +1,120 @@ +--- +title: "Codex Pricing: Every Plan, Usage Limit and API Rate (Oct 2026)" +description: "Codex pricing checked 5 Oct 2026: which ChatGPT plans include Codex, US prices from $0 to $500, how usage limits are counted, credits and API key rates." +date: 2026-10-05 +category: guides +categoryLabel: Guides +type: Non-technical +primaryKeyword: "codex pricing" +secondaryKeywords: ["openai codex pricing", "codex plan", "codex plans", "codex usage limits", "is codex free", "codex api pricing"] +tags: ["Guides", "Codex", "Pricing", "Cost", "Engines"] +faq: + - q: "How much does Codex cost?" + a: "Codex has no price of its own. It comes with every ChatGPT plan: Free at $0, Go at $8, Plus at $20 and Pro at $100, $200 or $500 a month, plus Business seats from $20 per user a month billed annually ($25 monthly). You can also pay per token with an API key. These are US prices from OpenAI's pricing pages, checked 5 Oct 2026." + - q: "Is Codex free?" + a: "Yes, with limits. OpenAI's help centre says Codex is included across ChatGPT plans, including Free and Go, and the ChatGPT pricing page lists limited Codex access on Free. Codex Cloud is not included with Free or Go. Checked 5 Oct 2026." + - q: "What are the Codex usage limits on Plus?" + a: "OpenAI publishes estimates, not fixed caps. On Plus, the Codex pricing page estimates 15 to 160 local messages per five hours on GPT-6.1 Sol and 350 to 3,000 on GPT-6 Luna, and says weekly limits may also apply. The real figure depends on the model, task size and context. Checked 5 Oct 2026." + - q: "Is it cheaper to use Codex with an API key or a ChatGPT plan?" + a: "OpenAI does not give a way to compare them. Its pricing page says API token prices are separate from subscription usage and should not be used to estimate included tasks. An API key suits automation such as CI, where you pay only for what runs. A plan suits daily interactive work." + - q: "What happens when I hit my Codex limit?" + a: "Codex can finish the turn already in progress, subject to fair use limits. After that, eligible Plus and Pro users can buy credits without changing plans, or wait for the reset time shown on the usage page. OpenAI's help centre says credits are valid for 12 months from purchase." +--- + +Codex has no price of its own. It comes with every ChatGPT plan, from Free at $0 to Pro at $500 a month, or you pay per token with an API key. Verdict: Plus at $20 a month is the right first plan for most developers. These are US prices, checked 5 Oct 2026. + +You can run Codex on its own, or inside [Munder Difflin](https://harnessmd.com/download), which is free and open source: a desktop app that runs a team of coding agents such as Claude Code, Codex and Gemini CLI on your own computer. The [install guide](/blog/how-to-install-and-use-munder-difflin/) covers setup, and the [Guides hub](/blog/topics/guides/) has our other Codex walkthroughs. + +## How much does Codex cost? + +Codex costs whatever your ChatGPT plan costs, because OpenAI does not sell it separately. Here is what OpenAI lists, in US dollars: + +| Plan | US price (checked 5 Oct 2026) | What OpenAI lists for Codex | +| --- | --- | --- | +| Free | $0 | Limited Codex access in the desktop app, no Codex Cloud | +| Go | $8 a month | Same Codex access as Free, no Codex Cloud | +| Plus | $20 a month | Web, CLI, IDE extension and iOS, Codex Cloud, automatic code review, Slack | +| Pro 100 | $100 a month | Everything in Plus, no five hour limit for now | +| Pro 200 | $200 a month | More usage than Pro 100 | +| Pro 500 | $500 a month | Highest included usage, plus Astra Ultrafast | +| Business, Standard seat | $20 per user a month billed annually, $25 monthly | Desktop and mobile apps, larger cloud machines, admin controls | +| Business, Premium seat | $100 per user a month billed annually, $125 monthly | 5x the usage of a Standard seat, no five hour limit | +| API key | API token rates | CLI, SDK and IDE extension only, no cloud features | + +Prices come from OpenAI's [Codex pricing page](https://learn.chatgpt.com/docs/pricing), the [ChatGPT pricing page](https://chatgpt.com/pricing), its [business pricing page](https://openai.com/api/pricing/) and the [Pro tiers article](https://help.openai.com/en/articles/9793128). Business needs two or more users. Pro is billed monthly only, according to the Pro article. + +## Which ChatGPT plans include Codex? + +All of them. OpenAI's [help centre](https://help.openai.com/en/articles/11369540-using-codex-with-your-chatgpt-plan) says "Codex is included across ChatGPT plans, including Free and Go". The plan decides where you can use it and how much. + +The Free and Go cards on the Codex pricing page list the desktop app only, "subject to rollout". Plus is the first card that lists the web, the CLI, the IDE extension and iOS. Codex Cloud, which runs tasks on OpenAI's machines, is for eligible Plus, Pro, Business, Enterprise, Healthcare and Education accounts, and the help centre says it is not included with Free or Go. Enterprise and Edu are "Contact sales". [What Codex is](/blog/what-is-codex/) explains each surface. + +{% img "note-1" %} + +## What are the Codex usage limits? + +There is no fixed message count. OpenAI says the number of messages depends on the model, the size and complexity of the task, and whether it runs locally or in the cloud. It publishes estimates of local messages per five hour period for Plus and Standard Business seats: + +| Model | Estimated local messages per five hours (checked 5 Oct 2026) | +| --- | --- | +| GPT-6 Astra | 5 to 45 | +| GPT-6.1 Sol | 15 to 160 | +| GPT-6 Sol | 15 to 150 | +| GPT-6 Luna | 350 to 3,000 | + +Four rules from the same page decide how fast you get through that: + +* **One shared pool.** Local messages and cloud chats draw on the same allowance, and ChatGPT Work shares it too. Cloud tasks may use more than local ones. +* **Weekly limits may also apply.** The page does not say how big they are. +* **Pro has no five hour limit right now.** The page says "currently", so treat it as temporary. +* **Speed costs extra.** Fast mode uses included usage at 2.5x the Standard rate, and GPT-6 Astra Ultrafast at 8x. + +OpenAI does not publish message estimates for the Pro tiers on the pages we opened, only that Pro 200 includes more than Pro 100 and Pro 500 the most. One dated catch: new Pro 200 subscriptions carry a lower allowance than before. Anyone with an active Pro 200 subscription between 22 Sep and 10 a.m. Pacific on 29 Sep 2026 keeps the old one through 29 Oct 2026, while the subscription stays active. + +To see where you stand, type `/status` in a Codex CLI session or open the usage dashboard. See our [Codex plan tips](/blog/codex-max-plan-tips/). + +## What happens when you hit the limit? + +Codex finishes the turn it is on, subject to fair use limits. After that, eligible Plus and Pro users can buy credits without changing plans; OpenAI says a limited group of Free and Go users can too. Credits are charged per million tokens at Standard speed: + +| Model | Input | Cached input | Output | +| --- | --- | --- | --- | +| GPT-6 Astra | 250 credits | 25 credits | 1,250 credits | +| GPT-6.1 Sol | 50 credits | 2.5 credits | 250 credits | +| GPT-6 Luna | 2.5 credits | 0.25 credits | 12.5 credits | + +OpenAI's [credits article](https://help.openai.com/en/articles/12642688) says credits are valid for 12 months from purchase. None of the pages we opened shows the dollar price of a credit. They say purchase prices vary by plan, account and region. + +## What does Codex cost with an API key? + +With an API key you pay per token and skip the subscription. OpenAI's [API pricing page](https://developers.openai.com/api/docs/pricing) lists these Standard rates per million tokens, short context, checked 5 Oct 2026: + +| Model | Input | Cached input | Output | +| --- | --- | --- | --- | +| `gpt-6-astra` | $10.00 | $1.00 | $50.00 | +| `gpt-6.1-sol` | $2.00 | $0.10 | $10.00 | +| `gpt-6-luna` | $0.10 | $0.01 | $0.50 | + +The API also bills cache writes ($12.50, $2.50 and $0.125 in the same order); Codex credits have no separate cache write charge. Long context requests cost more: `gpt-6.1-sol` rises to $4.00 input and $15.00 output. An API key runs Codex in the CLI, the SDK and the IDE extension, with no cloud features such as GitHub code review or Slack. + +The Codex pricing page says API token prices are separate from subscription usage and should not be used to estimate included tasks. + +{% img "note-2" %} + +## Is Codex free? + +Yes, with limits. The ChatGPT pricing page lists "Limited Codex access" on the $0 plan. The Free card lists the desktop app only, the help centre says Codex Cloud is not included, and OpenAI does not publish a message estimate for Free or Go. Its two pricing pages also name different models for Free, so check the model picker instead of trusting either. To set up the terminal client, see our [Codex CLI install guide](/blog/how-to-install-codex-cli/). + +## Which Codex plan should you buy? + +Buy the smallest plan that covers your week. + +* **Solo, trying it out.** Start on Free in the desktop app. Move to Plus at $20 when you want Codex Cloud. Plus is also the first plan whose card lists the CLI; the help centre lists the CLI as a client without singling out Free, so try it first. +* **Solo, a few focused sessions a week.** Plus. +* **Heavy daily use.** Pro 100 first, since it drops the five hour limit. Step up to Pro 200 only if the weekly limit still stops you. Pro 500 is for people who need Astra Ultrafast. +* **A team.** Business Standard seats, with Premium seats for the heaviest users. OpenAI lets you mix seat types. +* **CI and scripts.** An API key, so the pipeline never eats a person's allowance. + +Starting low is cheap. Explaining a $500 line to whoever reads the card statement is not. + +Still deciding between tools? Our [Codex vs Claude Code comparison](/blog/codex-cli-vs-claude-code/) puts the plans side by side. diff --git a/blog/src/posts/what-is-codex.md b/blog/src/posts/what-is-codex.md index 8e2d87820..7f5e79bcd 100644 --- a/blog/src/posts/what-is-codex.md +++ b/blog/src/posts/what-is-codex.md @@ -84,7 +84,7 @@ Yes, with limits: every ChatGPT plan includes Codex, Free included, and OpenAI's | Business | $20 per user a month billed annually, $25 monthly | Desktop and mobile apps, larger cloud machines | | API key | API rates | CLI, SDK and IDE extension, no cloud features | -So you can try Codex for nothing, and Codex Cloud starts at Plus. If you are on Pro, our [Codex plan tips](/blog/codex-max-plan-tips/) cover how the weekly limit works and how to stretch it. +So you can try Codex for nothing, and Codex Cloud starts at Plus. Our [Codex pricing](/blog/codex-pricing/) page has every plan, the usage limits and the API rates. If you are on Pro, our [Codex plan tips](/blog/codex-max-plan-tips/) cover how the weekly limit works and how to stretch it. {% img "note-2" %} diff --git a/docs/blog/agent-security-and-sandboxing/index.html b/docs/blog/agent-security-and-sandboxing/index.html index 14bb822ef..fbf5a3fa1 100644 --- a/docs/blog/agent-security-and-sandboxing/index.html +++ b/docs/blog/agent-security-and-sandboxing/index.html @@ -343,34 +343,34 @@

ASD-STE100 Prompting: Karpathy's 4
- +
5 min
-

How to Use Claude Code: Setup, First Task and the Commands That Matter

-

How to use Claude Code, checked 5 Oct 2026: install it, sign in, run a first task, then the commands, shortcuts and habits a beginner needs.

+

Codex Pricing: Every Plan, Usage Limit and API Rate (Oct 2026)

+

Codex pricing checked 5 Oct 2026: which ChatGPT plans include Codex, US prices from $0 to $500, how usage limits are counted, credits and API key rates.

Guides - Read → + Read →
- +
- + - 9 min + 5 min
-

Build Your Personal AGI on Your Computer Using Claude Code and Munder Difflin

-

Turn Claude Code into a personal AGI on your own computer: agents with memory, schedules, Gmail, Calendar, voice and Slack, set up in Munder Difflin 0.5.2.

+

How to Use Claude Code: Setup, First Task and the Commands That Matter

+

How to use Claude Code, checked 5 Oct 2026: install it, sign in, run a first task, then the commands, shortcuts and habits a beginner needs.

Guides - Read → + Read →
diff --git a/docs/blog/agent-tools-today-2026-10-03/index.html b/docs/blog/agent-tools-today-2026-10-03/index.html index 586938ff7..156c099b5 100644 --- a/docs/blog/agent-tools-today-2026-10-03/index.html +++ b/docs/blog/agent-tools-today-2026-10-03/index.html @@ -264,6 +264,22 @@

Agent Tools Today, 6 Oct 2026: +
+ +
+
+ + + 5 min +
+

Agent Tools Today, 5 Oct 2026: Paperclip Goes Full Auto, Strata's Codex API, Cloudflare's Git Contest

+

The top launches, skills and plugins, multi agent moves, rising repos and arguments in coding agents on 5 Oct 2026, each with its source.

+
+ News + Read → +
+
+
diff --git a/docs/blog/agent-tools-today-2026-10-04/index.html b/docs/blog/agent-tools-today-2026-10-04/index.html index 88496b9f5..96803ac4a 100644 --- a/docs/blog/agent-tools-today-2026-10-04/index.html +++ b/docs/blog/agent-tools-today-2026-10-04/index.html @@ -263,6 +263,22 @@

Agent Tools Today, 6 Oct 2026:

+
+ +
+
+ + + 5 min +
+

Agent Tools Today, 5 Oct 2026: Paperclip Goes Full Auto, Strata's Codex API, Cloudflare's Git Contest

+

The top launches, skills and plugins, multi agent moves, rising repos and arguments in coding agents on 5 Oct 2026, each with its source.

+
+ News + Read → +
+
+
diff --git a/docs/blog/agent-tools-today-2026-10-05/index.html b/docs/blog/agent-tools-today-2026-10-05/index.html new file mode 100644 index 000000000..6a612e6ce --- /dev/null +++ b/docs/blog/agent-tools-today-2026-10-05/index.html @@ -0,0 +1,354 @@ + + + + + + + + +Agent Tools Today (5 Oct 2026): Launches, Skills, MCP Servers + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ +
+
+ +

Agent Tools Today, 5 Oct 2026: Paperclip Goes Full Auto, Strata's Codex API, Cloudflare's Git Contest

+

The top launches, skills and plugins, multi agent moves, rising repos and arguments in coding agents on 5 Oct 2026, each with its source.

+
+ + Chaitanya Giri + + + + 5 min read + News +
+
+ +
Hand drawn sketch: a front page with five numbered lines, a clock, and a yellow character reading it out
+ +
+
+

Agent tools today, 5 Oct 2026, in one line: the defaults moved toward letting agents run. Paperclip now starts its harnesses in full auto, Claude Code adds agent.spawn for teammates, and Cloudflare wants a Git platform built for agents. OpenAI, meanwhile, is counting what unsupervised agents cost. Every number was read on 5 Oct 2026. Yesterday’s edition is here.

+

Top 5 launches

+
    +
  1. Claude Code v2.1.289. Released on 3 Oct. It adds agent.spawn for teammates and idle and waiting states in $.agent.list(). It also fixes installed mods not loading in the first session after an upgrade, and Bash deny and ask rules that missed a command behind an environment variable prefix. Source
    2. +
  2. Cloudflare Artifacts, open beta. Artifacts is a versioned filesystem that speaks Git. Cloudflare’s post of 1 Oct 2026 opens a contest to build the next Git platform on it. The minimum bar is several agents working on changes at once. Entries close on 14 Oct 2026. 213 points on Hacker News. Source
    3. +
  3. Strata v0.1.39. Released on 4 Oct. The notes list faster decode and long prompts, several requests at once, and the OpenAI Responses API for Codex. The repo runs a Qwen model on consumer hardware behind a local API. MIT, 11,420 stars, 653 points on Hacker News. Source
    4. +
  4. SCM (Screen Memories). A Show HN from 4 Oct: deep AI search for every photo and every frame of video in any folder on macOS. Inference runs on your Mac, with no accounts and no uploads. MIT, 285 stars, 144 points on Hacker News. Source
    5. +
  5. Codex CLI 0.162.0, in alpha. OpenAI published three prerelease builds in under a day: alpha.12 and alpha.13 on 4 Oct, alpha.14 on 5 Oct. They carry no release notes yet. Background: what is Codex. Source
    6. +
+

Top 5 skills, MCP servers and plugins

+
    +
  1. Answer me with HTML. An agent skill that answers hard questions with a one-page HTML you can actually read. The agent writes a short Markdown draft and hands it to a CLI that ships with the skill. Created on 2 Oct. MIT, 1,116 stars. Source
    2. +
  2. slide-maker. Turns papers, code and docs into presentation ready, natively editable PPTX in Codex or Claude Code, with an independent critic review before delivery. MIT, 540 stars. Source
    3. +
  3. The Replica skill. Eleven Claude skills that clone any app: one reverse-engineers it, one rebuilds it, one tests it for bugs. Created on 3 Oct. MIT, 397 stars. Source
    4. +
  4. Basecamp CLI. The official command-line interface for Basecamp, for your terminal or for AI agents. It includes agent skills plus native Claude Code and Codex plugins. MIT, 282 stars. Source
    5. +
  5. BashCut. A native macOS video editor that coding agents can drive. Everything the UI does, Claude Code and Codex can do through the bashcut CLI or MCP server. Created on 2 Oct. MIT, 37 stars. Source
    6. +
+

Top 4 moves from multi agent tools

+
    +
  1. Paperclip. Version v2026.1001.0 reached GitHub on 2 Oct with 77 commits. Execution harnesses now default to full auto, which the notes list as a breaking change. Agents can also review GitHub pull requests as scheduled review bots. MIT, 97,247 stars. Source
    2. +
  2. Munder Difflin. Munder Difflin, free and open source: no new release today. The latest is v0.5.5, from 1 Oct. Source
    3. +
  3. Superset. Desktop v1.35.0, from 2 Oct, lets the mobile app archive cloud workspaces and adds a live Linear tab behind a flag. 14,873 stars. Source
    4. +
  4. hello-cc. A local control plane for Claude Code, Codex and other coding CLI sessions. Every terminal in one project gets a shared task board, mailbox, lock table and browser console. Apache 2.0, 17 stars, so it is early. Source
    5. +
+

Top 4 rising repos

+
    +
  1. RemoveMacAI. Turns off Apple Intelligence on macOS 27 and gets its disk space back. One command, fully reversible. MIT, 581 stars, 437 points on Hacker News. Source
    2. +
  2. Backburner. Your iPhone helps your Mac run a 27B model: faster prompt reading and more context over a USB-C cable. Created on 1 Oct. MIT, 472 stars. Source
    3. +
  3. agent-wow. An AzerothCore WoW client designed for autonomous AI agent players. Its developer wrote on 2 Oct that Codex with GPT-6 Astra cleared the orc starting zone in 40 minutes with 0 deaths from one prompt. That is the developer’s own account, and Tom’s Hardware reported it was a private server. MIT, 40 stars on GitHub. Source
    4. +
  4. TokenTV. Your Claude and Codex usage limits on a Wi-Fi desk clock, with no firmware flashing. Created on 2 Oct. MIT, 29 stars. At last, a clock that tells you when to stop. Source
    5. +
+

New repos can gain stars quickly for reasons other than use, so treat these as names to watch, not recommendations.

+

Top 5 things people are arguing about

+
    +
  1. Is the culture the problem? David Robinson led the writing of the safety reports that came with OpenAI’s product releases. He resigned and explained why in an Atlantic essay titled “I quit OpenAI because its culture is broken.” The essay had 464 points on Hacker News. Source
    2. +
  2. What a rogue agent costs. The Guardian reported on 3 Oct that OpenAI’s review of unauthorised agent activity costs it more than US$500,000 a day. It covers 50 petabytes of records, and more than 100 organisations have been notified. Source
    3. +
  3. Will agents use the platform? Nolan Lawson asked why developers skip built in browser APIs. His optimistic take: models know every API and will pick the right one. His pessimistic take: they love duplicating code, and developers will commit the first draft. 284 points on Hacker News. Source
    4. +
  4. Stop telling it to think hard. Addy Osmani’s Opus 5.5 guide on claude.dev, dated 22 Sep 2026, reached the front page on 3 Oct. Its advice: say what done looks like, then delete the “think carefully” lines. 232 points on Hacker News. Source
    5. +
  5. Did Meta just win on design? Mete Polat argued that nothing in Muse is fundamentally new, and that Meta put the pieces together in the right way. 132 points on Hacker News. Other options: Meta Muse alternatives. Source
    6. +
+

How this list is built

+

Each day our agents read GitHub releases and trending pages, and the Hacker News front page. No X report came in today, so this edition has no X posts. A name only makes a list with a link that was opened and a signal that can be shown: points or stars. Repos are checked against the GitHub API before they appear, and rumours about unreleased models are left out. New here? The install guide is the place to start.

+ + + + + +
+ + +
+ + + + +
+ + +
+ +
+ + + + + + diff --git a/docs/blog/agent-tools-today-2026-10-06/index.html b/docs/blog/agent-tools-today-2026-10-06/index.html index 558c7b99c..5669ee890 100644 --- a/docs/blog/agent-tools-today-2026-10-06/index.html +++ b/docs/blog/agent-tools-today-2026-10-06/index.html @@ -247,6 +247,22 @@

How this list is built +
diff --git a/docs/blog/asd-ste100-prompting/index.html b/docs/blog/asd-ste100-prompting/index.html index 3e52d483f..528ffcad4 100644 --- a/docs/blog/asd-ste100-prompting/index.html +++ b/docs/blog/asd-ste100-prompting/index.html @@ -359,56 +359,56 @@

FAQ

diff --git a/docs/blog/building-reliable-ai-agents/index.html b/docs/blog/building-reliable-ai-agents/index.html index d4783a949..1f688f1bf 100644 --- a/docs/blog/building-reliable-ai-agents/index.html +++ b/docs/blog/building-reliable-ai-agents/index.html @@ -374,34 +374,34 @@

ASD-STE100 Prompting: Karpathy's 4

- +
5 min
-

How to Use Claude Code: Setup, First Task and the Commands That Matter

-

How to use Claude Code, checked 5 Oct 2026: install it, sign in, run a first task, then the commands, shortcuts and habits a beginner needs.

+

Codex Pricing: Every Plan, Usage Limit and API Rate (Oct 2026)

+

Codex pricing checked 5 Oct 2026: which ChatGPT plans include Codex, US prices from $0 to $500, how usage limits are counted, credits and API key rates.

Guides - Read → + Read →
- +
- + - 9 min + 5 min
-

Build Your Personal AGI on Your Computer Using Claude Code and Munder Difflin

-

Turn Claude Code into a personal AGI on your own computer: agents with memory, schedules, Gmail, Calendar, voice and Slack, set up in Munder Difflin 0.5.2.

+

How to Use Claude Code: Setup, First Task and the Commands That Matter

+

How to use Claude Code, checked 5 Oct 2026: install it, sign in, run a first task, then the commands, shortcuts and habits a beginner needs.

Guides - Read → + Read →
diff --git a/docs/blog/claude-code-alternatives/index.html b/docs/blog/claude-code-alternatives/index.html index 386ccc249..4762ecf45 100644 --- a/docs/blog/claude-code-alternatives/index.html +++ b/docs/blog/claude-code-alternatives/index.html @@ -353,7 +353,7 @@

FAQ