Tuesday, September 15, 2026

Creates architecture decision records so a team can document a choice, its context, and the trade-offs behind it instead of leaving design rationale scattered through chats and pull requests.

Useful when agent-built systems need an audit trail for decisions that will outlive the current implementation.

skills.sh · Engineering practice · High

Guides work through the Google Agents CLI, giving an agent a repeatable way to run the CLI rather than treating each task as a one-off command sequence.

Worth testing if your team wants a consistent harness around Google agent work instead of prompt-only operating habits.

skills.sh · Agent workflow · High

Adds a focused testing workflow for Vitest, helping an agent work with the JavaScript test runner while writing, running, and diagnosing tests.

A practical guardrail for code agents: tests catch the plausible-looking changes that miss the actual requirement.

skills.sh · Testing · High

Provides an Astro-specific skill so an agent can work with the web framework using its conventions rather than applying generic React or static-site assumptions.

Useful for teams shipping content-heavy sites where framework-specific context saves cleanup work later.

skills.sh · Web development · Medium

Frames a security-audit task for an agent, giving it a defined way to inspect an application or change set for security issues before work ships.

Good fit for an approval step before automated code changes reach production systems.

skills.sh · Security · High

Gathers live signals spanning power networks, energy markets, and data-center operations so an agent can answer infrastructure questions with a common operational view.

Relevant for teams operating compute-heavy workloads where power and grid constraints are becoming part of planning.

Smithery · Infrastructure intelligence · Medium

Checks whether a task should use local Ollama inference or cloud inference before execution, acting as a routing layer for model choice.

Useful for controlling cost, latency, and data exposure without forcing every workflow onto one inference path.

Smithery · Model routing · High

rlaope/oh-my-hermes

An all-in-one plugin for Hermes Agent that packages extra capabilities around the coding-agent environment.

MIT · 2,216 total

666ghj/MiroFish

A swarm-intelligence engine aimed at modeling and predicting outcomes, putting multi-agent simulation into a project you can inspect and run.

AGPL-3.0 · 73,404 total

tech-leads-club/agent-skills

A registry for professional agent skills that focuses on secure, validated entries rather than an uncurated pile of prompts.

No license · 6,163 total

microsoft/AI-Engineering-Coach

A Microsoft project for better agentic engineering, aimed at improving how teams build software with AI agents in the loop.

MIT · 4,165 total

NangoHQ/nango

An integration platform for products using AI, giving builders a way to connect external services without rebuilding each connector.

No license · 12,111 total

Andon Labs describes Pion as an agent intended to run a company autonomously. It is a concrete glimpse at how far founders want to push delegation beyond a single task or department.

HN

Amazon Science asks why ML research agents avoid overfitting, a question that matters when an automated research loop starts optimizing against its own narrow measurements.

HN

A practitioner documents moving large preprompts away from hosted models to self-hosted Ollama. The useful part is the operational friction that appears when prompt context meets local serving.

HN

Latent Space reports that xAI, OpenAI, and Anthropic have all cosigned the AEF-1 standard for third-party evaluators, putting more structure around external AI assessment.

Latent Space

A practitioner challenges memory-tool marketing based on retrieval latency, arguing that a low P50 figure alone says little about whether a voice agent feels useful.

R/AI_AGENTS

A team operating high-volume support asks where agents genuinely help in contact centers, a grounded prompt to separate useful automation from vendor demos.

R/AI_AGENTS

A builder shares an offline RAG setup across 27 technical books and 11,000 pages, with gold-set measurements of retrieval changes rather than just impressions.

R/RAG

A RAG debugging post traces confident wrong answers to the model ignoring part of the supplied context, even though retrieval itself looked healthy.

R/RAG

A developer publishes a comparison of GPT-5.6 Luna and GPT-6 Astra across 50 real pull requests and asks for scrutiny of the evaluation method.

R/CHATGPTCODING

Zero raised a 0.3m seed round backed by Lovable, Supercell, and Langdock founders, according to Sifted.

0.3m · Seed

Brussels-based Chift raised a €10.5m Series A to build a European financial-connectivity layer.

€10.5m · Series A

A paper proposing a living database and search engine for AI benchmarks and evaluation, aimed at researchers and developers working with LLMs and other AI systems.

September 15, 2026

Introduces ZGCM-1, an open 7B dense model trained for mathematical work and agentic search, making it the most directly operator-relevant model release in this set.

September 15, 2026

Studies post-training attention sparsification to reduce the quadratic cumulative attention cost by optimizing which context receives attention.

September 15, 2026