Mintlify’s self-updating documentation system uses two cooperating agents: a background agent that tunes the setup and runner agents that act when a trigger fires. Triggers can be schedules, code merges, or events in tools such as Jira; runners use skills to produce documentation changes, usually as pull requests, and notify reviewers. The background agent reviews run histories, integrations, platform analytics, and PR acceptance to adjust triggers, skills, and shared memory. For example, after code-merge-triggered CLI documentation changes were repeatedly rejected because the CLI ships continuously, it moved that work to a monthly release-notes schedule. The system has raised the internal acceptance rate for documentation PRs from 70% to 90%.
The central design choice is structured, git-backed memory rather than relying mainly on free-form notes. JSON schemas capture release details, stakeholders, review preferences, relevant discussions, style guidance, users, and integration data, making context more dependable and easier to evaluate. Mintlify built its own agent harness with tools for subagents, session forking, memory compaction, skills, and sandboxed shell commands; in offline tests it performed similarly to Pi while using 50% fewer tokens. Run logs and outputs are stored as JSON, with forked sessions recorded as a tree so complete histories can be reconstructed. The architecture also supports updates to customer setups: background agents receive instructions about schema or product changes, audit for regressions, and can decide which customers and contacts are appropriate for new features.
Summary generated by AI from the linked article. hn.today is not affiliated with Hacker News or Y Combinator.
Anthropic has restricted Claude from helping with video game modding, decompiling, and reverse-engineering, amid growing use of AI to alter games. The limits are broad enough to disrupt legitimate development and basic troubleshooting, and Anthropic has not publicly explained the restrictions. (mashable.com)
OpenAI’s GPT-6.1 Terra is a midrange model focused on visual reasoning, computer use, and technical communication, with faster computer-use performance than Astra and Sol. It costs $1 per million input tokens and $5 per million output tokens, positioning it between the cheaper Luna and pricier Sol. (simonwillison.net)
Pope Leo XIV argued that AI’s future depends on human choices and that systems should serve people, not be treated as persons. He called for human oversight of high-stakes decisions, steps to ban lethal autonomous weapons, and fair access to AI’s benefits. (garymarcus.substack.com)
Researcher Jiahao Shao says Anthropic blocked or downgraded his access after flagging his work as training a competitor. He says Claude only helped run jobs and analyze results, not generate training data. (twitter.com)
GPT-5.6 Sol completed a peaceful Factorio run in 13.5 hours of game time using a Python tool to control the game. It launched a rocket despite late-game factory problems, improving on an earlier GPT-5.5 attempt that failed to use conveyor belts. (ryanmadden.net)
A language model was post-trained on short Dwitter demos to generate plausible continuations of unfinished code and explore alternative visual effects. The project treats demo remixes as a kind of programming grammar, where tiny code changes can produce strikingly different results. (datavorous.com)
Today's best Hacker News stories, summarized and screenshotted, one email a day.