Coding Assistants

This plugin watches your Claude Code sessions, turns recurring patterns into native skills, and quietly archives the ones you stop using.

unlazy forces AI agents to prove their work through runnable acceptance gates instead of trusting their own exit reports.

Most browser agents chat at you; this one returns editable spreadsheets, slides, and sites you can actually use.

Because writing LinkedIn content that sounds human is tedious, this repo packages 11 prompt-engineered skills that let your AI agent draft, audit, and humanize posts—then wait for your approval before publishing.

It stops long-running agents from drifting by forcing them to verify progress through independent roles before every new round.

Moli is a standalone Rust headless browser that avoids continuous rendering by computing layout and paint only when an operation explicitly demands geometry or pixels.

zg indexes your workspace so both terminal users and AI agents can query by meaning or exact text without sending code to the cloud.

It gives AI agents hands-on access to local and remote machines without letting them wander off-leash or consuming a dedicated coding-agent quota.

It structures AI agents into a three-role math modeling team—modeler, coder, and writer—enforcing stage-gate quality checks before generating competition-ready Word or LaTeX papers.

Collie is a phone web UI that surfaces your Herdr agent herd over Tailscale, ranking agents by who needs you and turning prompts into tappable push notifications.

fx exists to provide a tiny, embeddable coding agent that behaves like a Unix shell rather than a bloated terminal IDE, with model-agnostic support for both local and cloud inference.

Cain is a real-world AI pentesting agent that enforces scope through engineering constraints rather than model self-discipline, with built-in coverage for AWS, Azure, GCP, and the Chinese clouds most tools ignore.

OpenContext is a personal knowledge store that plugs into the coding agent CLI you already use so your project context survives across repos, sessions, and tools.

Hyperresearch turns Claude Code into a tiered, multi-agent research pipeline that persists every source in a searchable, compounding vault.

67 game-dev skills and a router that auto-loads the right context so your AI agent stops guessing your engine.

This is a deliberately thin Python layer that hands an LLM raw macOS primitives and lets it write whatever app-specific glue it needs on the fly.

Webcmd learns a site's navigation once, then compiles it into deterministic CLI adapters so agents stop burning tokens rediscovering the same pages.

Because AI agents code fast but not necessarily well, these rule sets inject classic software engineering wisdom straight into their system prompts.

Because DeepSeek Harness treats every component as a swappable plugin, and this list tries to keep the resulting sprawl organized.

Phone Harness turns macOS iPhone Mirroring into a stateless bridge so an LLM can see, tap, and type on a real iPhone without jailbreaks, Xcode, or WebDriverAgent.