Claude Code plugin for AI-assisted development: spec-driven task-workflow with an independent verifier, mechanical enforcement gates, and pinned-library knowledge distillation
翻譯準備中
取自儲存庫 README
正在載入完整 README。 以下為已儲存的摘錄。
A plugin for Claude Code and Cursor, for AI-assisted development.
The agent writes a spec before it writes code. A second, memoryless agent audits the diff against that spec — never against the first agent's account of what it did. Commit-time hooks enforce both mechanically, rather than as prose in a doc nobody reads.
npx skills add tammai/bigin-skills installs the skills alone: no subagents (so no verifier and no routing tiers), no plugin hooks, and the ${CLAUDE_PLUGIN_ROOT} paths skills use to reach each other don't resolve. Use it only to try a single standalone skill.檢視完整 README ↗目前無法載入完整 README。請透過原文連結閱讀。 以下為已儲存的摘錄。
A plugin for Claude Code and Cursor, for AI-assisted development.
The agent writes a spec before it writes code. A second, memoryless agent audits the diff against that spec — never against the first agent's account of what it did. Commit-time hooks enforce both mechanically, rather than as prose in a doc nobody reads.
npx skills add tammai/bigin-skills installs the skills alone: no subagents (so no verifier and no routing tiers), no plugin hooks, and the ${CLAUDE_PLUGIN_ROOT} paths skills use to reach each other don't resolve. Use it only to try a single standalone skill.檢視完整 README ↗