feat: reset first-run experience for v0.2.1#600
Conversation
… placeholder sentinel - Default init creates 4 core squads (intelligence, research, product, company) with 14 agents - Add SYSTEM.md (Layer 0) and directives.md (Layer 3) to context cascade - BUSINESS_BRIEF.md has PLACEHOLDER sentinel — agents stop and ask user to edit instead of producing garbage - Upgrade all agent templates with structured output formats and quality rules - Add product squad (lead, scanner, worker) with depends_on intelligence/research - Rename create → add, remove list (use status), add --pack option to init - Clean repo root: remove changeset, husky, changelog, scripts, telemetry docs - Add AGENTS.md (vendor-neutral) and rewrite CLAUDE.md with cascade documentation - Update README with 4-step customization guide (Brief → Directives → Goals → Priorities) - Fix broken command references in company agents (squads feedback add → real commands) Co-Authored-By: Claude <noreply@anthropic.com>
Summary of ChangesHello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed! This pull request significantly refines the initial user experience and underlying agent framework for Highlights
Changelog
Activity
Using Gemini Code AssistThe full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips. Invoking Gemini You can request assistance from Gemini at any point by creating a comment using either
Customization To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a Limitations & Feedback Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here. You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension. Footnotes
|
…de files, sentinel Verifies init creates exactly the promised scaffolding: - 4 core squads (company, intelligence, product, research) - 14 agent definition files - SYSTEM.md, BUSINESS_BRIEF.md, directives.md cascade files - PLACEHOLDER sentinel in BUSINESS_BRIEF.md - CLAUDE.md and AGENTS.md at repo root Co-Authored-By: Claude <noreply@anthropic.com>
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Code Review
This pull request significantly improves the first-run experience and overall structure of the project. The cleanup of legacy files, simplification of the CLI, and introduction of more structured agent templates are excellent changes. The documentation in the README and other markdown files has been substantially improved, providing much clearer guidance for users and contributors. The addition of the product squad and the restructuring of the research squad are thoughtful improvements to the default agent workforce. The multi-provider support in the run command is also a great step forward.
| ```bash | ||
| npm install -g squads-cli | ||
| squads init | ||
| squads run research/analyst |
There was a problem hiding this comment.
The suggested command squads run research/analyst appears to be incorrect for the new squad structure. The research squad's lead agent is now named lead, while analyst is a worker agent. For a 'Get Started' example, running the lead agent is more appropriate as it orchestrates the squad's activities. This is also inconsistent with other parts of the codebase (like init.ts and README.md.template) which correctly suggest squads run research/lead. To ensure a smooth first-run experience for new users, I recommend updating this command.
| squads run research/analyst | |
| squads run research/lead |
… sentinel - --yes default: "A startup using AI agents to automate business operations" - Research focus: "AI agent and automation market" — produces interesting output on first run - PLACEHOLDER sentinel only appears when user skips description in interactive mode - Test updated: --yes brief has real content, no sentinel Co-Authored-By: Claude <noreply@anthropic.com>
…f, squads run - gh CLI: change from hard error to warning (not required for init/run) - gh CLI: fix install hint from `brew install gh` to https://cli.github.com - Remove duplicate Claude CLI check in runAuthChecks - Auto-commit scaffolding on init (agents need HEAD for worktrees) - Default business brief: "autonomous execution" + track big AI players - Getting Started step 3: `squads run` autopilot (replaces `squads dash`) - Add Docker first-run test environment - Version bump to 0.2.1 Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Claude <noreply@anthropic.com>
- Inline smoke test: check 4 core squads (not engineering/marketing/ops) - Inline smoke test: verify SYSTEM.md, directives.md, AGENTS.md - Inline smoke test: fix state file paths (research/lead not researcher) - Restore e2e-smoke.sh (was accidentally deleted) - Replace `squads list` with `squads status` in smoke script Co-Authored-By: Claude <noreply@anthropic.com>
1 participant
Summary
squads initcreates 4 core squads (intelligence, research, product, company) with 14 agents — matches READMEdepends_on: [intelligence, research]Test plan
npm run buildpassesnpm test— 1731/1731 tests passsquads init --yescreates all 4 squads, SYSTEM.md, directives.md, BUSINESS_BRIEF.md with sentinelsquads statusworks on fresh init (4 squads, 14 agents)🤖 Generated with Claude Code