Search

Canonical configs, fanned out to all your AI coding agents from the Commons, the canonical ~/.agents directory

Qwen3.8 27B is an open-weight dense vision-language model from Qwen. It is suited for coding, professional workflows, research, multimodal interaction, and long-running agent tasks, with flexible thinking that can be...

Thin Node/TypeScript projection of the shared Rust authority

Reusable agent-fleet machinery projected out of a production multi-harness agent workspace: skills, charters, policies, and the harness-sync engine that lands them in Claude, Codex, Gemini, Cursor, and friends.

Agnostic CLI to scaffold and enforce Spec-Driven Development (SDD) with AI agents

No description available

One-shot scaffolder for product monorepos: a walking skeleton, a verify gate, a path to production, and a repository set up for coding agents.

Pareto is a multimodal composite model built for research, coding, and agentic workflows, while delivering frontier-level performance across a broad range of general-purpose tasks.

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....

Shift every LLM call to the cheapest model that can handle it. Routing decided by TypeSafe Jev in ~300 ms. No training data. Policy in plain YAML.

Multi-agent evaluation and reasoning trace extraction for scientific workflows

Jev-powered smart tool-prune context engine for Hermes Agent (any OpenAI-style System One endpoint)

`createLocalSession` from `@agenticos-dev/bot-testkit/local-session` wraps the facet runtime with a bounded JSON request interface. Supply trusted modules, explicit browser-callable `allowedMethods`, host-only `seed` calls, and exact HTTP loopback `origin

Local contract-5 project context, evidence, decisions, work, and continuity

Deterministic did:key Ed25519 invocation signer (CapabilityAgent) derived from a seed or a salted secret, for the browser, Node.js, and React Native.

Headless coding-agent runtime for unattended, host-scheduled use: zero-dependency, dual-protocol streaming, tool loop, context compaction, checkpoint/resume. The CLI is a verifier/debugger.

Evolution of context-first-cli: manage the Context-First methodology across any project AND orchestrate dynamic, ephemeral AI agents from specs. All agent orchestration lives in .md command templates; the Node layer only scaffolds and manages.

Yuanjian — a zero-dependency static malware triage engine for AI agents: MD5/SHA1/SHA256 hashes, magic-type detection, Shannon entropy, printable strings (ASCII/UTF-16LE) with URL/domain/IP/email/command/path/base64 classification, and PE/ELF header parsi

Fork of @deepseek-ai/dsh-agent-instructions that delivers the workspace instruction baseline as system-prompt sections, keeping the upstream mid-session reminder path.

The SpyneJS Generative Grammar — what AI agents read to write well-formed SpyneJS. Content-only package: cards (the operations), records (the constructions), the mental model, and the trees that route any task to them; versioned with framework releases.