MikeTrendsTrends right now
AI: reposAPI

search

agentic coding

Trends

  1. 1
    Caveman-style tool cuts AI coding agent token use●JuliusBrussee/caveman⬢github209just now

    Developer JuliusBrussee has released an open-source project called caveman, a skill and proxy for coding agents written in Go. It prompts AI coding assistants to write in a stripped-down, 'caveman' style, which the project says cuts token consumption by around 65 percent while keeping the code output intact. The project is drawing attention among developers looking to reduce costs on popular coding agents.

  2. 2
    Superpowers framework for AI coding agents gains traction●obra/superpowers⬢github556just now

    Developer obra has released Superpowers, an open-source agentic skills framework and software development methodology designed to make AI coding agents work more effectively. The project, written in Shell, describes itself as a methodology that 'works,' and is drawing attention as developers look for practical ways to structure and improve how autonomous coding agents handle real software projects.

  3. 3
    Ponytail tool makes AI agents code as lazily as possible●DietrichGebert/ponytail⬢github1.4Kjust now

    A JavaScript project called Ponytail by developer Dietrich Gebert is circulating among coders. It is designed to make AI coding agents behave like 'the laziest senior dev in the room', pushing them to write as little code as possible. The pitch is that the best code is the code you never wrote, a philosophy that is drawing attention and engagement from developers tracking new AI coding tools.

  4. 4
    Matt Pocock releases engineer-focused AI agent skills repository●mattpocock/skills⬢github955just now

    TypeScript educator Matt Pocock has published a repository called 'skills' on GitHub, described as 'Skills for Real Engineers', drawn from his personal .agents directory. The collection packages configurations he uses with AI coding agents. It quickly climbed GitHub's trending ranks, drawing attention from developers interested in how experienced engineers structure AI-assisted workflows.

  5. 5
    Context-mode tool targets AI coding agent memory limits●mksglu/context-mode⬢github282just now

    A new open-source project called context-mode, written in TypeScript, aims to optimise how AI coding agents use their context windows. It claims to sandbox tool output with a 98% size reduction, persist session memory between tasks, and enforce routing across 17 platforms using MCP and hooks. Developers are sharing it as a practical response to the common problem of agents running out of usable context during long sessions.

  6. 6
    Open-source marketing skills released for Claude Code agents●coreyhaines31/marketingskills⬢github140just now

    Developer Corey Haines published a JavaScript toolkit called marketingskills, which adds marketing capabilities to Claude Code and other AI agents. The collection covers conversion rate optimisation, copywriting, SEO, analytics and growth engineering. It is gaining traction on GitHub, where early users see it as a way to turn coding assistants into marketing assistants.

  7. 7
    CodeGraph project offers pre-indexed knowledge graphs for AI coding tools●colbymchenry/codegraph⬢github98just now

    A developer-published open-source project called CodeGraph is drawing attention. It builds a pre-indexed knowledge graph of a codebase that automatically syncs as code changes, and works with AI coding assistants including Claude Code, Codex, Gemini, Cursor, OpenCode, AntiGravity, Kiro, Copilot and Hermes Agent. The pitch is fewer tokens and tool calls, with everything running locally.

  8. 8
    Open-source model routing targets top coding-agent performance●Show HN: Open-source model routing for coding agents at Astra-level performanceYhnEnvironmentOceans1157 min ago

    A developer has released an open-source tool that routes requests between AI models for coding agents, claiming performance on par with Astra-level systems. The launch invites developers to test whether smart routing across cheaper models can match premium single-model results. Early discussion centers on benchmarks, reliability, and whether routing adds real gains for coding workflows.

  9. 9
    Developers Delete Unit Tests as AI Coding Agents Spread●Developers Delete Unit Tests as AI Agents Reshape Coding𝕏xSE60513 min ago

    Software developers are reportedly removing unit tests from their codebases as AI coding agents take on more of the programming work. The practice is drawing debate, with some arguing tests become unnecessary when AI generates and verifies code, while others warn deleting tests undermines safety nets and long-term code quality.

  10. 10
    Ask HN: Is anybody producing good code with coding agents?●YhnScienceBiology2519 min ago

    A discussion thread on Hacker News asks whether developers are actually producing good code with AI coding agents. The question invites engineers to share firsthand experiences on whether these tools deliver production-quality results or mostly generate code that needs heavy review and rework.

  11. 11
    Graphene launches as data analysis toolkit for coding agents●Show HN: Graphene – Data analysis toolkit for your coding agentYhnEnvironmentOceans718 min ago

    A new open-source project called Graphene has been released, offering a data analysis toolkit designed to work with AI coding agents such as coding assistants that write and run code. The toolkit is available on GitHub, where developers can try it and contribute. Early reactions on Hacker News are modest but curious, with the community discussing whether agent-focused data tools will become a standard part of developer workflows.

Repos