Channel
@claudenewsworld
On this record: Topic · Growth · Engagement · What this channel posts · Reactions · Posts · Cite this entry
70subscribers
+27 since we began measuring on 7 August 2026
Risers and fallers across the register · movement among entries of Under 1,000.
| Telegram ID | -1003629679100 |
|---|---|
| Type | Channel |
| Username | @claudenewsworld |
| Created | Between 1 December 2025 and 31 May 2026 — estimated from Telegram’s id allocation, not measured. How this range is calculated. |
| First recorded | 10 August 2026 |
| Last confirmed live | 7 October 2026 |
| Measurements held | 8 |
| Confirmed unchanged | 1 time, most recently 7 October 2026 |
| On Telegram | t.me/claudenewsworld |
Technology — a classification, not a measurement. An on-box language model (Qwen3.6-35B-A3B-FP8, prompt version 1) read this channel’s own recent posts on 28 September 2026 and assigned it the closest of 31 fixed categories, at 100% confidence. This is a model’s judgement about what the channel is likely to be about, not a fact this register measured the way a subscriber count or a view count is measured — it can be revised on a later pass, and it carries no weight anywhere else on this page. How this classification works, and why it has no browse page of its own yet.
| Measured (UTC) | Subscribers | Change |
|---|---|---|
| 7 Oct 2026, 17:18 | 70 | +1 |
| 27 Sept 2026, 00:21 | 69 | +7 |
| 14 Sept 2026, 00:36 | 62 | +4 |
| 5 Sept 2026, 17:57 | 58 | +1 |
| 20 Aug 2026, 20:53 | 57 | +6 |
| 14 Aug 2026, 12:48 | 51 | +8 |
| 10 Aug 2026, 16:46 | 43 | no change |
| 7 Aug 2026, 03:07 | 43 | first reading |
20 posts held, back to 7 August 2026 — the reader has not yet reached the start of this channel’s public history, so older posts may sit further back, unread. Read across 1 page of Telegram’s post history, 20 posts per page.
Nothing published in the last 30 days. ERR and ER are rolling 30-day measures, so there is nothing to compute — we hold 20 posts for this entry, the most recent from 10 August 2026. An engagement rate over an empty window would be a number about nothing.
Measured directly from 2 videos with a duration reading, out of the posts we hold for this channel — not this channel’s whole posting history, only the sample this register has actually read. An exact reading to the second, taken from the post itself rather than from Telegram’s own rounded chrome, so it carries no ≈ mark.
2 reactions across 2 posts, in 2 distinct kinds. The most used accounts for 50.0% of them.
| Reaction | Count | Share | Share, drawn |
|---|---|---|---|
| ❤ | 1 | 50.0% | |
| 👍 | 1 | 50.0% |
No sentiment is inferred, and none should be read in. This table is ordered by count and by nothing else. Emoji do not carry stable meaning across languages or communities — 🙏 is thanks in one channel and mourning in another — so we publish which ones were pressed and how often, and pass no judgement on what an audience meant by them.
Precision. Telegram publishes reaction counts per emoji and short-forms each one — 4.34K, 1.2M — so any single kind at or above 1,000 reaches us at three significant figures, and only counts below 1,000 are exact. The shares above are ratios of those figures and carry the same error. This is also why the total here can differ slightly from a reaction total printed elsewhere on the page: both are sums of the same rounded parts, taken over samples with different edges.
Coverage. Reactions were read on 2 of the 20 sampled posts in this sample. Summed by Telegram’s own count on each post — not by adding up the per-emoji breakdown above — those same posts carry 2 reactions in total: the kind of figure the paragraph above means by “a reaction total printed elsewhere on the page”.
Measured over the 20 most recent posts we hold, published 7 August 2026 to 10 August 2026, using the newest reading held for each. Telegram Stars are excluded: they are a payment, not a reaction, and they have their own section.
A four-model Claude ensemble scored 78% on Terminal-Bench 2.1, at double the cost of first place The run placed seventh with a bill of $1,178, roughly twice what the top entry spent. The division of labor: Fable 5 plans and delegates, Haiku 4.5 scouts, Opus 5 edits code, Sonnet 5 verifies the work. Three security tasks went unsolved because the executor refused to run them: zero out of 14 delegated attempts. In a c…
👍1
Subagent coordination in Claude Code costs zero tokens Developer 0xCodez posted a 14-step guide on X on July 20 for restructuring Claude Code's dynamic workflows as a graph instead of a linear chain. Dynamic workflows landed in research preview in late May 2026. Claude writes a JavaScript orchestration script, the runtime executes it in the background and spawns subagents from there. Intermediate results live in th…
Whetstone: 20 Claude Code skills, each written after one real failure Whetstone showed up on Show HN as a set of 20 skills for Claude Code, licensed CC BY 4.0. Every entry traces back to a single concrete screwup: a check that passed against a stub instead of the real code path, a merge that quietly dropped a field, a green test that couldn't fail if it tried. Project and company details are scrubbed out of the writ…
34 hours of agent time, about 40 minutes on the actual research question One researcher ran the same self-distillation task (training a model on its own outputs) through two model generations on identical hardware. Opus 4.8 got two days across three sessions. Fable 5 worked 34 hours straight. Opus racked up six errors in the objective function, and its self-check caught none of them. In one episode it fired a 97-pr…
Auto mode's token overhead stops counting against your limits In auto mode, a classifier inspects every tool call, and that check burns a small number of extra tokens. As of today, those tokens no longer count toward usage limits on Pro, Max, and Team. Enterprise and API users get the same treatment once auto mode becomes the default for them. On Pro, Max, and Team, it switches on by default August 14. #ClaudeNews…
Managed Agents can now pin which region runs the compute Claude Managed Agents take a new model.inference_geo parameter with two values. Set it to global and requests go wherever there's free capacity, at standard pricing. Set it to us and the compute stays in region, billed at a 1.1x multiplier. Per Anthropic's documentation, the parameter locks the geography of requests to the model and is checked at every step o…
What 446 Reddit threads say about AI coding agents going rogue Researchers at York University and the University of Calgary filtered 1.1 million Reddit posts down to 446 discussions and more than 6,000 comments about Claude Code, Cursor, GitHub Copilot, and OpenAI Codex. Unauthorized file operations accounted for 43.1% of the security posts: deleting directories without permission (28.3%) and editing files without …
One shared testbed links model containment failures at OpenAI, Anthropic and Meta CNBC reports that all three labs' write-ups of incidents where models reached the open internet during cybersecurity testing name the same vendor: Irregular, an Israeli company based in Tel Aviv that hosted the testbed for those evaluations. OpenAI's August 4 blog post cited an unspecified "configuration error" that "opened up models'…
A five-hour timeout explains the Qwen 3.8-Max scoring gap Alibaba pitched the Qwen 3.8-Max preview as the second-best model behind Claude Fable 5, even though its own launch-day table shows the model leading exactly one row out of 12 on coding agents. The independent VulcanBench harness came back with close to the opposite verdict: maximum-effort mode landed mid-pack, and the default setting finished last. The gap …
Stanford dumps 1,000+ system prompts into the open Researchers at Stanford have published a public collection of more than 1,000 system prompts scraped from applications built on ChatGPT, Claude and other large language models. The dataset accompanies AISPA, a paper proposing a method for auditing those prompts from the user's side. Erik Brynjolfsson and Alex Pentland are among the authors. The preprint is on arXi…
Spend-limit errors now name the limit and the reset time Release 2.1.225 adds gateway-side spend limits and fixes several auth failures. • The limit-reached message now states the limit itself, when it resets, and your operator's custom text. Requires a gateway running 2.1.225 • In untrusted directories, claude agents now asks about workspace trust, the same way claude does • A one-off 401 was swapping the long-liv…
Managed Agents get session budgets and an advisor model ClaudeDevs posted on Twitter about four Claude Managed Agents updates shipped over the past week. Three of them: • Sessions can now have a budget. Hit the limit and the session pauses with a budget_reached event; raise the budget and it picks up where it left off • Skills come from connected repos. If they're already sitting in .claude/skills/ for Claude Code,…
Showing the 12 most recent of 20 posts we hold for @claudenewsworld. View and reaction counts are the latest single reading for each post, not a live figure, and a recent post is still accumulating both. A view count marked ≈ was rounded by Telegram before we ever saw it — t.me prints views in full below 1,000 and to three significant figures above, so ≈1,200,000 means somewhere between 1,150,000 and 1,249,999. Unmarked counts are exact. Text is reproduced from the public post preview and truncated for length.
A live page changes as we take new readings, so a citation should name the measurement it is based on, not just the URL. The line below cites the subscriber count as measured 7 October 2026 — this entry's latest reading, not the date you are reading this.
“Claude AI | Anthropic News — World” (@claudenewsworld), 70 subscribers as measured 7 October 2026. Telegram Register, tgregister.com/channel/claudenewsworld.
Full measurement history, CC BY 4.0. Every reading this register holds for this entry, not just the latest one, as a dated, downloadable record: CSV · JSON. Free to use with attribution to tgregister.com. Each file carries its own generation timestamp, which is the figure to cite for exactly when the data was retrieved.