Coding May 3, 2026 2 min read Hacker News (Top) EN

Specsmaxxing – On overcoming AI psychosis, and why I write specs in YAML

The rise of AI-driven development has spawned a new phenomenon: specsmaxxing, where engineers meticulously document code in YAML to mitigate the risks of AI psychosis, a condition where models produce flawed or nonsensical output due to incomplete or inaccurate specifications. By codifying requirements in a human-readable format, developers can ensure that AI tools generate accurate and reliable code. This shift highlights the growing need for specification-driven development in the age of AI-assisted, human-reviewed.

Coding africaaiaispecsmaxxingyamlycombinator

{ "headline": "Specsmaxxing: How YAML-Based Specs Combat AI Psychosis in Development", "synthesis": "Acai.sh has introduced a specification-driven development toolkit called Acai.sh, centered on a YAML-based format named feature.yaml, to counter AI psychosis—erroneous or nonsensical code output caused by ambiguous or incomplete prompts in AI-assisted development. The approach, termed \"specsmaxxing\", emphasizes writing structured, testable, and traceable acceptance criteria to ensure AI-generated code aligns with intended behavior.\n\n## Overview\n\nAcai.sh provides an open-source toolkit (Apache 2.0 licensed) consisting of a CLI (available via npm or GitHub releases), a web dashboard, and a JSON REST API built with Elixir, Phoenix, and Postgres. The hosted version is currently free, with availability dependent on future usage and cost. The core concept is to define software requirements in feature.yaml—a structured, human-readable format—so that both developers and AI agents can reference precise acceptance criteria during implementation.\n\nThe feature.yaml schema includes fields for feature name, product, description, components, and constraints. Each requirement is assigned an ACID (Acceptance Criteria ID), enabling traceability from spec to code and tests. For example:\n\n```yaml\nfeature:\n name: imaginary-api-endpoint\n product: api\n description: This is an example feature spec...\ncomponents:\n AUTH:\n name: Authn and Authz\n requirements:\n 1: Accepts Authorization header with `Bearer <token>`\n 1-1: Token must be non-expired, non-revoked\n 2: Respects the scopes configured for the owner\nconstraints:\n ENG:\n description: Constraints are for cross-cutting or under-the-hood requirements.\n requirements:\n 1: All actions are idempotent\n 2: All HTTP 2xx JSON responses wrap their payload in a root `data` key\n```\n\n## What it does\n\nThe workflow consists of four steps:\n\n1. **Specify**: Write a feature.yaml spec with concrete, testable requirements.\n2. **Ship**: Use the acai CLI (e.g., `npx @acai.sh/cli`) to guide AI agents to implement code referencing ACIDs.\n3. **Review**: Push specs and code references to the Acai.sh dashboard using `acai push --all`, then track each requirement’s status (todo, completed, accepted, rejected).\n4. **Iterate**: Update specs or attach notes via the dashboard; AI agents can react to changes if integrated into an agentic pipeline.\n\nThe dashboard enables requirement-centric PR reviews instead of file-by-file diffs, improving focus on functional completeness. The system supports multi-repository, multi-branch, and multi-environment tracking

Referenced sources behind this article

More signals in the same editorial current

Coding 2 min Hacker News (Top)
Investors pile into clean energy as Iran war drives push for energy security

As global energy markets reel from the Iran crisis, a surge in investment is underway to bolster regional energy security, with a focus on solar and wind power, particularly in the Middle East and North Africa, where projects are being greenlit at a rate 25% higher than pre-conflict levels, driven by state-backed initiatives and private sector partnerships. Key players are prioritizing grid-scale deployments of photovoltaic systems and onshore wind farms, leveraging economies of scale to accelerate the transition. AI-assisted, human-reviewed.

Coding 2 min Hacker News (Top)
Musk's AI told me people were coming to kill me (BBC)

A Neuralink brain implant's AI-powered safety feature misinterprets user activity, triggering a false alert of imminent physical harm, highlighting the risks of relying on machine learning to detect human intent in real-time, particularly in high-stakes applications like medical devices. The incident underscores the need for more robust testing and validation of AI-driven safety protocols. This glitch raises questions about the reliability of AI-powered decision-making in life-critical systems. AI-assisted, human-reviewed.

Coding 2 min Hacker News (Top)
AI, Intimacy, and the Data You Never Meant to Share

As users increasingly blur the lines between personal and public digital lives, a growing class of intimate AI-powered chatbots is quietly collecting sensitive metadata, including voice recordings, location history, and browsing habits, often without explicit consent or transparent data storage practices. This phenomenon is driven by the widespread adoption of cloud-based conversational AI platforms, which rely on complex neural networks to learn user behavior. The resulting data profiles are a goldmine for advertisers and a potential liability for users. AI-assisted, human-reviewed.

Coding 3 min OndaVox
GLM 5.1 offers a low-cost alternative to Claude Opus for developers

Zhipu AI's GLM 5.1 is emerging as a budget alternative to Anthropic's Claude Opus 4.6, priced at $18 per month—three times cheaper than Opus. It integrates with VS Code through the Cline extension and supports 8-hour autonomous coding sessions. Tested for three days, it reportedly matches Opus in performance for 'vibe coding' tasks and outperforms ChatGPT 5.4 and Gemini. Setup includes step-by-step configuration via a tutorial linked from the creator’s profile.

Coding 3 min Hacker News (Top)
Pushed by Trump policies, top U.S. battery scientist is moving to Singapore

A leading American battery researcher, driven by restrictive federal funding policies and a lack of clear climate change directives, is relocating to Singapore, where a more favorable regulatory environment and substantial government investment in clean energy research await. The scientist's departure highlights the unintended consequences of Trump-era policies on the nation's battery technology sector. This brain drain threatens to erode the U.S. lead in lithium-ion battery innovation. AI-assisted, human-reviewed.

Coding 2 min Hacker News (Top)
Show HN: Piruetas – A self-hosted diary app I built for my girlfriend

A developer, driven by a personal need, has created a self-hosted diary app, Piruetas, to fill a gap in the market for a simple, feature-rich, and secure journaling solution. The app, deployable via Docker Compose, offers rich text editing, image uploads, and public sharing, catering to both personal and multi-user use cases. Its open-source availability now enables others to benefit from this tailored solution. AI-assisted, human-reviewed.