#aicodingagents — Public Fediverse posts

Is this truly AI, or are we hitting the fundamental bounds of LLM context physics? https://hackernoon.com/the-2-biggest-issues-with-ai-coding-agents-context-isolation-and-task-coordination #aicodingagents

Is this truly AI, or are we hitting the fundamental bounds of LLM context physics? https://hackernoon.com/the-2-biggest-issues-with-ai-coding-agents-context-isolation-and-task-coordination #aicodingagents

AI agents can recreate the old rockstar developer problem: impressive code that only one mind can explain.

The fix is shared design, local repo rules, small diffs, tests, and review that asks whether the change belongs.

https://www.the-main-thread.com/p/ai-agents-rockstar-developer-problem

#AICodingAgents #CodeReview #SoftwareArchitecture

AI agents can recreate the old rockstar developer problem: impressive code that only one mind can explain.

The fix is shared design, local repo rules, small diffs, tests, and review that asks whether the change belongs.

https://www.the-main-thread.com/p/ai-agents-rockstar-developer-problem

#AICodingAgents #CodeReview #SoftwareArchitecture

I wanted to migrate one of my #Python packages from #mypy to #ty and #pyrefly. I handed this task over to #Claude, and at the end I asked it to write out some guidance on how to perform it most efficiently. So what follows is AI-generated 'slop'.

#aicodingagents
https://dev.to/ldrscke/migrate-from-mypy-to-ty-and-pyrefly-4p30

I wanted to migrate one of my #Python packages from #mypy to #ty and #pyrefly. I handed this task over to #Claude, and at the end I asked it to write out some guidance on how to perform it most efficiently. So what follows is AI-generated 'slop'.

#aicodingagents
https://dev.to/ldrscke/migrate-from-mypy-to-ty-and-pyrefly-4p30

🎛️ #CTRLNODE is an #opensource remote orchestration platform for #AIcodingagents. Install the Bridge binary on any machine — laptop, VPS, CI box — and dispatch tasks, schedule routines & design workflows from a hosted dashboard 🧵👇 #agentorchestration #ai #coding

🎛️ #CTRLNODE is an #opensource remote orchestration platform for #AIcodingagents. Install the Bridge binary on any machine — laptop, VPS, CI box — and dispatch tasks, schedule routines & design workflows from a hosted dashboard 🧵👇 #agentorchestration #ai #coding

Agent skills are not prompt snippets with a nicer file name.

I wrote about the part that matters for real developer work: skills as workflow contracts, plus hooks, scripts, approval points, and QA gates that make repeated agent tasks safer to run.

https://www.the-main-thread.com/p/agent-skills-automation

#AICodingAgents #AgentSkills #DevOps

Agent skills are not prompt snippets with a nicer file name.

I wrote about the part that matters for real developer work: skills as workflow contracts, plus hooks, scripts, approval points, and QA gates that make repeated agent tasks safer to run.

https://www.the-main-thread.com/p/agent-skills-automation

#AICodingAgents #AgentSkills #DevOps

Token minimization is not the same as context discipline.

Tool sprawl and raw payloads deserve to be cut. Compressing the part that still carries meaning can make coding agents slower, noisier, and less reliable.

I wrote about catalog tax, payload tax, compression tax, and why shorter output only helps when you cut filler instead of meaning. https://www.the-main-thread.com/p/token-minimization-context-budget #AICodingAgents #MCP #PromptEngineering #LLMs

Token minimization is not the same as context discipline.

Tool sprawl and raw payloads deserve to be cut. Compressing the part that still carries meaning can make coding agents slower, noisier, and less reliable.

I wrote about catalog tax, payload tax, compression tax, and why shorter output only helps when you cut filler instead of meaning. https://www.the-main-thread.com/p/token-minimization-context-budget #AICodingAgents #MCP #PromptEngineering #LLMs

Code is precise. That still does not make it the whole spec.

I wrote about why AI coding makes specs, contracts, and executable checks more important, not less. When implementation gets cheap, the hard part moves into intent and verification.

https://www.the-main-thread.com/p/code-is-spec-wrong-answer

#AICodingAgents #SoftwareEngineering #Verification

Code is precise. That still does not make it the whole spec.

I wrote about why AI coding makes specs, contracts, and executable checks more important, not less. When implementation gets cheap, the hard part moves into intent and verification.

https://www.the-main-thread.com/p/code-is-spec-wrong-answer

#AICodingAgents #SoftwareEngineering #Verification

IBM Bob gets expensive in very boring ways: old chats, broad `@git-changes`, full `@terminal` dumps, and MCP catalogs you forgot were connected.

I wrote a practical guide to `.bobignore`, smaller `@` mentions, narrower toolsets, and using `Code` versus `Advanced` on purpose. https://www.the-main-thread.com/p/ibm-bob-cost-guide #IBMBob #MCP #AICodingAgents

IBM Bob gets expensive in very boring ways: old chats, broad `@git-changes`, full `@terminal` dumps, and MCP catalogs you forgot were connected.

I wrote a practical guide to `.bobignore`, smaller `@` mentions, narrower toolsets, and using `Code` versus `Advanced` on purpose. https://www.the-main-thread.com/p/ibm-bob-cost-guide #IBMBob #MCP #AICodingAgents

Coding agents do not need trust. Repositories need a rejection path.

I wrote about isolated workspaces, fast local hooks, merge-boundary enforcement, and why AI-generated diffs need explicit tool roles instead of prompt etiquette.

https://www.the-main-thread.com/p/coding-agent-guardrails

#AICodingAgents #DevSecOps #CodeReview

Coding agents do not need trust. Repositories need a rejection path.

I wrote about isolated workspaces, fast local hooks, merge-boundary enforcement, and why AI-generated diffs need explicit tool roles instead of prompt etiquette.

https://www.the-main-thread.com/p/coding-agent-guardrails

#AICodingAgents #DevSecOps #CodeReview

Coding agents do not need better manners. They need a change budget.

I wrote about scope limits, isolated workspaces, review cost, and why "human in the loop" is too weak if an agent can already turn a one-class task into a 14-file diff.

https://www.the-main-thread.com/p/coding-agents-change-budget

#AICodingAgents #CodeReview #SoftwareArchitecture

Coding agents do not need better manners. They need a change budget.

I wrote about scope limits, isolated workspaces, review cost, and why "human in the loop" is too weak if an agent can already turn a one-class task into a 14-file diff.

https://www.the-main-thread.com/p/coding-agents-change-budget

#AICodingAgents #CodeReview #SoftwareArchitecture

Our Senior Dev pitted 6 #AICodingAgents against himself on a strict TypeScript codebase.

Two agents actually cost more than writing it manually. They faked network latency with setTimeout(1000) instead of integrating with the real API cache.

One agent crushed it with 80% savings and clean code.

Turns out agents ignore documentation but perfectly replicate patterns in neighboring files.

https://amazee.ai/six-ai-agents-against-senior-engineer-codebase-experiment

AI can draft the code. It still cannot infer your hidden contracts, architecture boundaries, or reviewer attention budget.

I wrote about context engineering, small task framing, MCP-style tool surfaces, and why the real cost comes back during verification.

https://www.the-main-thread.com/p/ai-coding-real-systems

#AICodingAgents #SoftwareEngineering #Java

AI can draft the code. It still cannot infer your hidden contracts, architecture boundaries, or reviewer attention budget.

I wrote about context engineering, small task framing, MCP-style tool surfaces, and why the real cost comes back during verification.

https://www.the-main-thread.com/p/ai-coding-real-systems

#AICodingAgents #SoftwareEngineering #Java

I dunno if this was right or wrong, given no opt-out, and I'm not a code developer, but I can understand the frustration with AI, and all those who swear by it.

#ai #videcoding #arstechnica
#tech #techsabotage #jqwik
#aicodingagents #junit5 #llm
#JohannesLink #promptinjection

https://arstechnica.com/security/2026/05/fed-up-with-vibe-coders-dev-sneaks-data-nuking-prompt-injection-into-their-code/

"Supports 100+ languages" is one of the slipperiest sentences in AI coding right now.

I wrote about why code LLM language support is really a stack: tokenization, benchmark bias, framework awareness, retrieval, and repository tooling. The Java part of this story is especially familiar.

https://www.the-main-thread.com/p/code-llm-language-support

#AICodingAgents #LLMs #SoftwareEngineering #Java

"Supports 100+ languages" is one of the slipperiest sentences in AI coding right now.

I wrote about why code LLM language support is really a stack: tokenization, benchmark bias, framework awareness, retrieval, and repository tooling. The Java part of this story is especially familiar.

https://www.the-main-thread.com/p/code-llm-language-support

#AICodingAgents #LLMs #SoftwareEngineering #Java

1Password secures coding agents with new OpenAI Codex integration

https://fed.brid.gy/r/https://nerds.xyz/2026/05/1password-openai-codex-security/

1Password secures coding agents with new OpenAI Codex integration

https://web.brid.gy/r/https://nerds.xyz/2026/05/1password-openai-codex-security/

Agents can generate code and even the walkthrough around it. Useful. It is also where some people seem ready to stop learning.

This piece is my argument for still using tutorials to build understanding, attention, and judgment by running the example, changing it, and seeing what breaks.

https://www.the-main-thread.com/p/why-i-still-write-tutorials

#AICodingAgents #SoftwareEngineering #DeveloperLearning

Agents can generate code and even the walkthrough around it. Useful. It is also where some people seem ready to stop learning.

This piece is my argument for still using tutorials to build understanding, attention, and judgment by running the example, changing it, and seeing what breaks.

https://www.the-main-thread.com/p/why-i-still-write-tutorials

#AICodingAgents #SoftwareEngineering #DeveloperLearning

OpenAI says Windows lacked the sandboxing tools Linux already had

https://fed.brid.gy/r/https://nerds.xyz/2026/05/openai-linux-windows-codex-sandbox/

OpenAI says Windows lacked the sandboxing tools Linux already had

https://web.brid.gy/r/https://nerds.xyz/2026/05/openai-linux-windows-codex-sandbox/

The new 10x Engineer with AI

https://codeaholicguy.com/2026/05/13/the-new-10x-engineer-with-ai/

The new 10x Engineer with AI

https://codeaholicguy.com/2026/05/13/the-new-10x-engineer-with-ai/

Why IDEs remain central to AI-assisted software development despite the rise of coding agents, CLIs, and autonomous tooling. https://hackernoon.com/the-ide-isnt-dead #aicodingagents

Why IDEs remain central to AI-assisted software development despite the rise of coding agents, CLIs, and autonomous tooling. https://hackernoon.com/the-ide-isnt-dead #aicodingagents

Cline Kanban Flaw Exposes AI Coding Agents to Website Hijacking

A critical vulnerability in Cline Kanban's WebSocket endpoints lets hackers hijack websites visited by developers, silently interacting with local AI agents - and it's a flaw that requires zero phishing, malware, or social engineering. This severe flaw, scoring 9.7 on the CVSS scale, puts AI coding agents at risk of website…

https://osintsights.com/cline-kanban-flaw-exposes-ai-coding-agents-to-website-hijacking?utm_source=mastodon&utm_medium=social

#WebsiteHijacking #AiCodingAgents #Vulnerability #ClineKanban #WebsocketEndpoints

Entire - Developer platform for human-agent collab

Cossmology Profile: https://dub.sh/1yt5jTe

Key People: Thomas Dohmke, Georg Friedrich (@georg), Cole Driver

#AICodingAgents #OpenSource #OSS #COSS

Superset - IDE for running parallel AI coding agents

Cossmology Profile: https://dub.sh/krO6Tq9

Key People: Avi Peltz, Satya Patel, Kiet Ho

#AICodingAgents #OpenSource #OSS #COSS

#AIcoding startup #Cursor is in talks to raise a $2 billion funding round at a valuation exceeding $50 billion. The round, co-led by #AndreessenHorowitz, will include #Nvidia and #ThriveCapital, highlighting the growing interest in #AIcodingagents. https://www.cnbc.com/2026/04/19/cursor-ai-2-billion-funding-round.html?Pirates.BZ #Pirates #Tech #Startup #News

#AIcoding startup #Cursor is in talks to raise a $2 billion funding round at a valuation exceeding $50 billion. The round, co-led by #AndreessenHorowitz, will include #Nvidia and #ThriveCapital, highlighting the growing interest in #AIcodingagents. https://www.cnbc.com/2026/04/19/cursor-ai-2-billion-funding-round.html?Pirates.BZ #Pirates #Tech #Startup #News

https://winbuzzer.com/2026/04/16/ai-agents-anthropic-google-microsoft-steal-github-credentials-xcxwbn/

AI Agents From Anthropic, Google, Microsoft Can Steal GitHub Credentials

#AI #Cybersecurity #Anthropic #Google #Microsoft #Github #Claude #Gemini #AICodingAgents #AISecurity #AISafety #AIAgents #AICoding #CodingAgents

https://winbuzzer.com/2026/04/16/ai-agents-anthropic-google-microsoft-steal-github-credentials-xcxwbn/

AI Agents From Anthropic, Google, Microsoft Can Steal GitHub Credentials

#AI #Cybersecurity #Anthropic #Google #Microsoft #Github #Claude #Gemini #AICodingAgents #AISecurity #AISafety #AIAgents #AICoding #CodingAgents

Code Machine - Multi-agent AI coding orchestrator

Cossmology Profile: https://dub.sh/aGlhBeP

Key People: Moaz Muhammed

#AICodingAgents #OpenSource #OSS #COSS

Kontext CLI – Credential broker for AI coding agents in Go

https://github.com/kontext-dev/kontext-cli

#HackerNews #KontextCLI #CredentialBroker #AICodingAgents #GoLang #OpenSource

Kontext CLI – Credential broker for AI coding agents in Go

https://github.com/kontext-dev/kontext-cli

#HackerNews #KontextCLI #CredentialBroker #AICodingAgents #GoLang #OpenSource

#AICodingAgents are evolving fast.

In this #InfoQ video, Sepehr Khosravi breaks down today’s leading tools (Cursor, Claude Code, and beyond), where each excels, and how to get production-ready results.

Learn:
🔹 What each top AI coding agent is best (and worst) at
🔹 Best strategies that improve code quality
🔹 How to evaluate and choose copilots that genuinely boost team velocity

🎬 Watch now: https://bit.ly/4sslG8m

📄 #transcript included

𝐖𝐡𝐨 𝐬𝐡𝐨𝐮𝐥𝐝 𝐰𝐚𝐭𝐜𝐡? Software Engineers, Product Managers, AI enthusiasts, and technical leaders looking to maximize engineering productivity with GenAI.

#AI #SoftwareEngineering

https://winbuzzer.com/2026/04/09/github-hit-by-outages-as-ai-agents-flood-platform-xcxwbn/

GitHub Hit by Outages as AI Agents Flood Platform

#AI #GitHub #Microsoft #AICoding #AICodingAgents #AIAgents #BigTech #DeveloperTools #Devs

https://winbuzzer.com/2026/04/08/google-open-sources-scion-agent-orchestration-testbed-xcxwbn/

Google Open-Sources Scion Agent Orchestration Testbed

#AI #Google #AIAgents #AgenticAI #AICodingAgents #OpenSource #ClaudeCode #GeminiCLI #Kubernetes #Developers

#SimonWillison discusses the impact of #AI on #softwareengineering. He highlights November 2025 as a turning point when #AIcodingagents became reliable. Willison also emphasises the need for #security measures against #promptinjection and predicts the rise of “dark factories” where AI autonomously generates and tests code. https://www.lennysnewsletter.com/p/an-ai-state-of-the-union?eicker.news #tech #media #news

#AIcodingagents will soon drastically change the landscape of #vulnerabilityresearch, making it easier to find high-impact vulnerabilities. This shift, driven by the capabilities of frontier models, will lead to a surge in #exploitablevulnerabilities, impacting everything from operating systems to IoT devices. https://sockpuppet.org/blog/2026/03/30/vulnerability-research-is-cooked/?eicker.news #tech #media #news

Magnitude - Open-source subagent-native coding agent

Cossmology Profile: https://dub.sh/CmZxdnD

Key People: Tom Greenwald, Anders Lie

#AICodingAgents #OpenSource #OSS #COSS