August 4, 2026
Bot Bosses or Bureaucracy?
Agent skills that bring team coding standards to Claude Code and Codex
This tool promises team order for AI coding, but commenters are already side-eyeing the chaos
TLDR: ADLC Team Skills wants to make AI coding assistants follow one shared team playbook instead of everyone improvising alone. Commenters are torn between calling it much-needed discipline and clowning it as bloated, confusing, and possibly more work than the problem it solves.
A new open-source project called ADLC Team Skills is pitching itself as the cure for AI coding anarchy. The big idea: instead of every developer secretly training their chatbot with private little tricks, the whole team shares one set of rules, goals, examples, and standards. In plain English, it’s trying to stop the “everyone lets the bot do whatever” era before companies drown in messy code, confusing changes, and finger-pointing.
But the real show is in the comments, where the crowd instantly split into “finally, some adult supervision” and “this is way too much homework for a coding tool.” One early supporter said their team already built something similar because AI assistants kept “drifting” off course, which gives the project some real-world credibility. Then came the skeptics, who basically asked: is this helpful structure, or just a deluxe machine for burning more time and money? One commenter dryly predicted the likely outcome would be more token burn, more review work, and slower tasks — then twisted the knife with, “Who knows, maybe that’s a good thing.”
And yes, the community brought the memes. One person complained the README was so confusing they couldn’t even tell how the thing works. Another delivered the cleanest tabloid-worthy jab of the thread: “Write your README or ME won’t READ it.” Brutal. Add in a one-word drive-by review — “Slop” — and suddenly this launch feels less like a calm product debut and more like a comment-section cage match over whether AI coding needs rules… or a translator.
Key Points
- •The article presents ADLC Team Skills as an open-source system for enforcing shared engineering standards across AI coding agents.
- •It pairs a skills repository with a companion directives repository that stores version-controlled team context such as constitutions, rules, personas, indexes, and manifests.
- •The quickstart uses `adlc-skills-cli` or `npx skills add`, and the article says the setup works with Agent Skills-compatible tools including Claude Code, Codex, OpenCode, Cursor, and GitHub Copilot.
- •The `adlc-skills-cli` wrapper can generate slash commands and wire `session_start` hooks via `.events.json`, while `mission-brief` can auto-discover skills from multiple repositories.
- •The setup flow includes `team-boot` and `team-setup`, with options to scaffold a new `team-ai-directives` repository or connect an existing one, and the system is organized around four team workflow pillars.