From deep-think
Use when the user conversationally asks to add reasoning rules, constraints, or anti-patterns to their deep-think configuration. Triggers on phrases like "remember to always check X", "add a rule about Y", "from now on check Z during reasoning", "never do X when thinking through problems".
How this skill is triggered — by the user, by Claude, or both
Slash command
/deep-think:manage-constraintsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
The user wants to add, remove, or view constraints that guide deep-think reasoning.
The user wants to add, remove, or view constraints that guide deep-think reasoning.
Constraints map to .deep-think.json in the current project directory:
practices.rules — things to always check during reasoningpractices.antiPatterns — patterns to avoid and flag during reflection.deep-think.json from the current project directory.
.deep-think.json back to disk.If unclear, ask the user which category fits better.
npx claudepluginhub bis-code/mcp-deep-think --plugin deep-thinkExtract conventions and implicit knowledge from session data into .claude/rules/local/ files. USE WHEN repeated patterns are found across sessions.
Interviews the user to capture a rule into `.chief/_rules/` or scaffolds the rules directory. Use when the user has a rule in mind and wants to write it proactively.
Guides creation of .claude/rules/ files for path-scoped project conventions using TDD workflow: RED (test gaps), GREEN (write rule), REFACTOR (optimize). Triggers on 'add rule', 'create convention', 'scope guideline'.