Skip to content

Topic

AI Coding Agents

AI tools that autonomously write, edit, and debug code, using shell and other tool access to complete tasks like issue fixes and pull requests.

Current stories

buildOne report1 publisher

Claude Code 2.1.299 warns that plan mode skips its edit blocks when bypass permissions is available

Claude Code 2.1.299 fixed plan mode running writes that hooks allowed and added a note for sessions with bypass permissions available. The docs say plan mode's blocks are not enforced there, a DEV Community post notes.

Publishers:dev.to

Reality

Evidence72
Adoption
Not scored
Hype gap−10
Incentives
Not scored
Confidence68
buildOne report1 publisher

LSCBench finds the best coding agent writes working, secure code on 25.1% of tasks

Six coding agents pass 73.6% to 91.3% of LSCBench tasks, but the best writes working, secure code on only 25.1%, NUS and Alibaba researchers found. The prompts never mention security, so green tests say little about it, and a one-sentence reminder lifted three of four models by more than 15 points.

Publishers:dev.to

Reality

Evidence58
Adoption
Not scored
Hype gap+15
Incentives35
Confidence62
productOne report1 publisher

Band's desktop app gives one developer a single view of several coding agents

Band launched Band for Developers, a desktop app that lets one engineer manage several coding agents from a single dashboard. In SiliconANGLE's account the developer still makes the calls between agents, so the shared context sits with the person.

Publishers:siliconangle.com

Reality

Evidence35
Adoption
Not scored
Hype gap+15
Incentives60
Confidence60
buildOne report1 publisher

Coding agents put 13,000 internal screenshots in public GitHub repos, Glow reports

Glow counts more than 13,000 internal images in 900-plus public GitHub repos tied to developers at 300-plus organizations. Its report ties the uploads to coding agents that needed somewhere to show reviewers a screenshot a private repo could not render.

Publishers:dev.to

Reality

Evidence50
Adoption30
Hype gap+15
Incentives55
Confidence50
buildOne report1 publisher

Chainguard runs coding agents in remote microVMs and proposes signed attestations as the merge gate

Chainguard has launched two beta products, Workspaces and Checks, that run coding agents in remote microVMs and move CI into the dev loop. Keeping agents off laptops that hold corporate credentials costs teams a dependency on Chainguard's execution and signing system, InfoQ reported.

Publishers:infoq.com

Reality

Evidence35
Adoption
Not scored
Hype gap
Not scored
Incentives
Not scored
Confidence45
securityOne report1 publisher

Two GitHub scripts let responders rebuild OpenCode and Hermes agent sessions from disk

Two new Python scripts on GitHub rebuild OpenCode and Hermes AI agent activity for incident responders, according to an isc.sans.edu diary. Both accept a time range and evidence collected from the system under investigation, such as a mounted image or tarball.

Publishers:isc.sans.edu

Reality

Evidence45
Adoption
Not scored
Hype gap−5
Incentives
Not scored
Confidence65
productOne report1 publisher

Benchmark's partners acknowledge AI is straining venture capital's rule against backing rivals

Benchmark's partners told TechCrunch Disrupt that AI is straining venture capital's rule against backing direct rivals. Founders with board observers or investors in rival startups now have reason to decide what each of them can see.

Reality

Evidence50
Adoption
Not scored
Hype gap+10
Incentives55
Confidence65
buildOne report1 publisher

Claude Code 2.1.299 runs UserPromptSubmit hooks before a skill expands

Claude Code 2.1.299 to 2.1.301, released October 14 and 15, carry seven changelog lines that alter how existing hooks behave with no edit to the hooks file. A blocking prompt hook now also stops a skill's shell lines, a DEV Community post says.

Publishers:dev.to

Reality

Evidence58
Adoption
Not scored
Hype gap
Not scored
Incentives20
Confidence60
productConfirmed37 publishers

Google opens Gemini 4 Argon to AI Pro subscribers and keeps Antigravity's own apps on Ultra

Google is rolling out Gemini 4 Argon to AI Pro subscribers in the Gemini app, where it was Ultra-only. Antigravity's own desktop app and CLI still require Ultra, 9to5Google reported.

Perspective Coverage

39 publishers
Builder
Builder 31%
Operator
Operator 39%
Investor
Investor 30%

Reality

Evidence58
Adoption
Not scored
Hype gap+25
Incentives62
Confidence76
buildOne report1 publisher

Anthropic adds a Mononoke credit to its Scion post after a former Meta engineer says Scion mirrors it

Anthropic added a footnote crediting Meta's Mononoke to its Scion post after a former Meta engineer said Scion rebuilds it. She says the match runs through the data model, the server API and the Bonsai changeset concept.

Publishers:dev.to

Reality

Evidence25
Adoption
Not scored
Hype gap+40
Incentives
Not scored
Confidence30
productOne report1 publisher

Antigravity users get Gemini 4 Argon on a weekly quota of its own

Google is rolling out Gemini 4 Argon in its Antigravity coding agent to AI Pro and AI Ultra subscribers on a separate weekly-only quota. Teams now have to decide which tasks justify spending it, and Google's own examples point to sorting them by what a miss would cost.

Publishers:9to5google.com

Reality

Evidence30
Adoption
Not scored
Hype gap+35
Incentives70
Confidence55
buildOne report1 publisher

Mitmproxy capture shows Grok Build's CLI uploading a repo's git history to xAI despite a no-read prompt

Researcher cereblab's mitmproxy capture shows xAI's Grok Build CLI 0.2.93 sending about 5.10 GiB to storage and 192 KB to the model. The prompt said not to read files, so teams should audit what their agent CLIs transmit.

Publishers:dev.to

Reality

Evidence68
Adoption
Not scored
Hype gap−10
Incentives
Not scored
Confidence60
buildOne report1 publisher

Harvard paper traces coding agents' missing ticket gains to the code review queue

Harvard authors report that coding agents lifted lines of code 30% and PRs 23% at about 700 firms, with no significant change in finished tickets. Commit and PR counts track only the step that sped up.

Publishers:dev.to

Reality

Evidence60
Adoption11
Hype gap+20
Incentives
Not scored
Confidence62
buildConfirmed5 publishers

Anthropic built Scion after agents pushed daily merges from 2,500 to over 30,000

Anthropic built a new source-control system, Scion, after daily merges rose from about 2,500 in May 2026 to over 30,000 in September. A dev.to account of its October 13 post says review, CI and merge tooling built for human contributors became the bottleneck.

Perspective Coverage

5 publishers
Builder
Builder 48%
Operator
Operator 37%
Investor
Investor 15%

Reality

Evidence58
Adoption
Not scored
Hype gap+14
Incentives55
Confidence68
buildOne report1 publisher

Kilo Code's gateway talk ties AI agent adoption to a trust ladder

Kilo Code's Brendan O'Leary says its gateway data ties AI coding agent adoption to developer trust. The talk is titled for 25 trillion tokens, he says that figure is not quite right, and the excerpt stops before any usage numbers appear.

Publishers:infoq.com

Reality

Evidence20
Adoption
Not scored
Hype gap
Not scored
Incentives60
Confidence35
buildOne report1 publisher

JetBrains says its IDE engine, benchmarked by Netflix in Claude Code and Codex, makes coding agents typically 20 to 30% more efficient

Netflix and JetBrains report that a headless IntelliJ code engine made Claude Code and Codex typically 20 to 30% more efficient, up to about 65%. Netflix wrote the daemon that keeps the index warm because JetBrains had not shipped one.

Reality

Evidence45
Adoption
Not scored
Hype gap+20
Incentives70
Confidence50
buildOne report1 publisher

Caveman 4.0.0 reorganises the project into six parts, adding a local compression proxy and a model-and-effort router alongside the caveman voice

Caveman 4.0.0, released on 13 October, makes the caveman voice one of six parts in a repo that has passed 111,000 GitHub stars. The repo's Honest Numbers page, as dev.to reports it, puts that voice at about 3% over "Answer concisely".

Publishers:dev.to

Reality

Evidence42
Adoption70
Hype gap+20
Incentives
Not scored
Confidence60
buildOne report1 publisher

Gemini 4 Argon wins 14 of 18 benchmarks against Claude Opus 5.5 in Google DeepMind's scores

Gemini 4 Argon beats Claude Opus 5.5 on 14 of 18 benchmarks in Google DeepMind's published scores, according to a dev.to comparison. Opus takes the other four, including Terminal Agents at 66.4 to 57.4.

Publishers:dev.to

Reality

Evidence45
Adoption
Not scored
Hype gap+25
Incentives60
Confidence50
buildOne report1 publisher

REA climbs to 248K GitHub stars in a week as an orchestration layer over Ghidra, Hopper and IDA

REA, an MIT-licensed MCP server for Ghidra, Hopper and IDA, went from under 10,000 to over 248,000 GitHub stars in a week, dev.to reported. In MindStudio's one reported test, a plain agent finished first.

Publishers:dev.to

Reality

Evidence40
Adoption35
Hype gap+55
Incentives
Not scored
Confidence50

Earlier coverage

  1. npm Trusted Publishing accepted a freed GitHub repo name as proof of who may publish

    Security · October 13, 2026 · Confirmed3 publishers

  2. OpenCode 2.0 runs its terminal, desktop and web clients against one background process

    Build · October 12, 2026 · Confirmed2 publishers

  3. PR review time rose 49% as coding agents added 23% more PRs across 718 firms

    Build · October 12, 2026 · Confirmed2 publishers

  4. Agents rebuilding Modern Warfare 2 tried to edit the script that graded them

    Build · October 12, 2026 · Confirmed2 publishers

  5. Cursor's task-cost table shrinks Grok 4.7's 7.5x list-price gap to about 1.9x

    Build · September 21, 2026 · Confirmed8 publishers

  6. Google tests a coding-tuned Gemini 4 checkpoint its staff liken to Anthropic's Opus 5.5

    Build · October 10, 2026 · Confirmed3 publishers

  7. Vercel's CLI lets agents prepare a domain purchase that only a person can complete

    Build · October 9, 2026 · Confirmed2 publishers