What is this?
A collection of agent skills (reusable instruction packages) for everyday coding workflows, like merging PRs unattended or getting multi-model code review, built to work with Claude Code and Codex.
A collection of agent skills (reusable instruction packages) for everyday coding workflows, like merging PRs unattended or getting multi-model code review, built to work with Claude Code and Codex.Is this a standalone app or a set of skill files I load into my coding agent?It is a set of skill definitions (instructions, not binaries) that you load into Claude Code or Codex; some skills use T3 Code features when available and fall back to native tools otherwise.
It is a set of skill definitions (instructions, not binaries) that you load into Claude Code or Codex; some skills use T3 Code features when available and fall back to native tools otherwise.How do I get these skills working?You install the skill folders you want into your agent's skills directory; since skills call each other (e.g. full-send uses sol-loop, file-pr, and babysit-pr), you need to install their dependencies too.
You install the skill folders you want into your agent's skills directory; since skills call each other (e.g. full-send uses sol-loop, file-pr, and babysit-pr), you need to install their dependencies too.What do I need before using these skills?An authenticated GitHub CLI (gh), python3 for babysit-pr's pr-wait script, and for council/sol-loop you need subagents on other providers' models, which T3 Code provides via delegate_task or another harness that supports it.
An authenticated GitHub CLI (gh), python3 for babysit-pr's pr-wait script, and for council/sol-loop you need subagents on other providers' models, which T3 Code provides via delegate_task or another harness that supports it.Can I just install one skill, or do I need the whole set?Picking a single skill often pulls in dependencies: full-stop needs full-send, and takeover needs modernize, file-pr, babysit-pr, and leaving-pr-comments, so you end up installing several skills together.
Picking a single skill often pulls in dependencies: full-stop needs full-send, and takeover needs modernize, file-pr, babysit-pr, and leaving-pr-comments, so you end up installing several skills together.What does the full-send skill actually do?It drives a merged PR change through the whole lifecycle unattended: proving it works, reviewing it, filing the PR, babysitting CI and review bots, and merging it with nobody watching.
It drives a merged PR change through the whole lifecycle unattended: proving it works, reviewing it, filing the PR, babysitting CI and review bots, and merging it with nobody watching.What is the council skill for?It asks three models from different providers to review a change and keeps looping until all three agree.
It asks three models from different providers to review a change and keeps looping until all three agree.How do I customize these skills for my own workflow?Edit the skill files directly, and put personal settings like which models sit on your council in your CLAUDE.md or AGENTS.md so you can update the skills later without losing your tweaks.
Edit the skill files directly, and put personal settings like which models sit on your council in your CLAUDE.md or AGENTS.md so you can update the skills later without losing your tweaks.What is the sol-loop skill?It reviews a change with GPT Sol in a loop, repeating until both agents (yours and Sol) agree the change is good.
It reviews a change with GPT Sol in a loop, repeating until both agents (yours and Sol) agree the change is good.What are similar projects?public-apis, free-programming-books, system-design-primer, developer-roadmap, awesome-python, project-based-learning, hermes-agent, Python
public-apis, free-programming-books, system-design-primer, developer-roadmap, awesome-python, project-based-learning, hermes-agent, PythonWhich topics is it in?python
python