rule-creator
Creates and manages Claude Code rules in .claude/rules/ with the Incorrect/Correct template, classifies the input before writing, and decides scope, location, and frontmatter from signals in the request. Supports create, list, edit, extract from oversized CLAUDE.md, delete, and refusal with redirect when the input belongs to a skill or hook. Use this skill whenever the user defines a coding convention, team standard, or constraint Claude should enforce — even when the user does not explicitly say "rule". Also use for scoping a rule to specific paths, splitting a growing CLAUDE.md into rule files, or auditing existing rules. Triggers: "create rule", "add rule", "new rule", "coding standard", "team convention", "list rules", "edit rule", "extract rules from CLAUDE.md", "delete rule". Not for procedural workflows, lifecycle hooks, or one-off task instructions.
What this skill does
# Rule Creator
Creates well-scoped rules in `.claude/rules/` and manages the rule set.
## Workflow
```
trigger --> dispatch --> classify --> context --> render --> write
| |
v v
list/edit refuse (procedural / lifecycle / one-off)
extract/del
```
Dispatch picks the mode from verb signals. Create runs the classifier
and project context check before rendering the template. Other modes
skip classification.
## Triggers and dispatch
| Signal in input | Mode | Load |
|-----------------|------|------|
| "create / add / new rule", "convention", "standard", or a declarative description with no verb | create | [classify-and-context.md](references/classify-and-context.md), then [rule-format.md](references/rule-format.md) |
| "list / show rules", "what rules exist" | list | [modes.md](references/modes.md) |
| "edit / update / change rule X" | edit | [modes.md](references/modes.md), [rule-format.md](references/rule-format.md) |
| "extract / split / move from CLAUDE.md", "CLAUDE.md is too big" | extract | [modes.md](references/modes.md) |
| "delete / remove rule X" | delete | [modes.md](references/modes.md) |
Scope is project only: rules land in `.claude/rules/`. User-level rules
(`~/.claude/rules/`) are out of scope.
## Create gates (run in order)
1. **Classify input.** Procedural multi-step → refuse and recommend
authoring a skill instead. Lifecycle event → refuse and recommend
a hook. One-off task → refuse, suggest doing it directly.
Declarative convention → proceed. See
[classify-and-context.md](references/classify-and-context.md).
2. **Context check.** Stack mismatch, duplicate topic, contradiction
with CLAUDE.md, or rule too vague to verify → flag and ask before
writing. Same reference.
3. **Scope decision.** Path signals (extension, directory, framework
name) → path-scoped `paths:` frontmatter. Otherwise global. Same
reference.
4. **Render.** Apply the Incorrect/Correct template strictly. See
[rule-format.md](references/rule-format.md).
5. **Verifiability checklist.** Action verb? Specific tool or syntax?
Reviewer can say "violated"? Fail any → rewrite before saving.
6. **Write.** New topic → new file. Existing topic without conflict →
append H2. Existing topic with conflict → ask user.
## Guidelines
- Default scope to path-scoped when the input names an extension, a
directory, or a framework; global only for universal conventions
- One topic per file; multiple rules in the same topic become H2
sections inside that file
- Filenames are kebab-case, descriptive nouns (`testing.md`,
`api-design.md`, not `rules.md` or `misc.md`)
- Sanitize the filename: lowercase, ASCII, hyphens only
- Refuse procedural input — recommend authoring a skill instead of
compressing the steps into a weak rule
## Anti-Pattern: Rule as Workflow
A rule that reads like a procedure ("first do A, then B, then C")
fails as a rule. Claude reads rules as standing constraints, not
playbooks. Multi-step content belongs in a skill. When the input is
sequenced and conditional, refuse and recommend authoring a skill
instead of compressing the steps into bullets.
## Anti-Pattern: Vague Conventions
"Format code properly" and "write good tests" cannot be checked. The
verifiability gate exists to catch these. If the rule has no action
verb, no specific tool or syntax, and no observable behavior, it is
decoration. Rewrite it concretely or drop it.
Related in AI Agents
skill-development
IncludedComprehensive meta-skill for creating, managing, validating, auditing, and distributing Claude Code skills and slash commands (unified in v2.1.3+). Provides skill templates, creation workflows, validation patterns, audit checklists, naming conventions, YAML frontmatter guidance, progressive disclosure examples, and best practices lookup. Use when creating new skills, validating existing skills, auditing skill quality, understanding skill architecture, needing skill templates, learning about YAML frontmatter requirements, progressive disclosure patterns, tool restrictions (allowed-tools), skill composition, skill naming conventions, troubleshooting skill activation issues, creating custom slash commands, configuring command frontmatter, using command arguments ($ARGUMENTS, $1, $2), bash execution in commands, file references in commands, command namespacing, plugin commands, MCP slash commands, Skill tool configuration, or deciding between skills vs slash commands. Delegates to docs-management skill for official documentation.
reprompter
IncludedTransform messy prompts into well-structured, effective prompts — single or multi-agent. Use when: "reprompt", "reprompt this", "clean up this prompt", "structure my prompt", rough text needing XML tags and best practices, "reprompter teams", "repromptception", "run with quality", "smart run", "smart agents", multi-agent tasks, audits, parallel work, anything going to agent teams. Don't use when: simple Q&A, pure chat, immediate execution-only tasks. See "Don't Use When" section for details. Outputs: Structured XML/Markdown prompt, quality score (before/after), optional team brief + per-agent sub-prompts, agent team output files. Success criteria: Single mode quality score ≥ 7/10; Repromptception per-agent prompt quality score 8+/10; all required sections present, actionable and specific.
adaptive-compaction
IncludedAdaptive add-on policy and recovery layer that decides WHEN to compact, prune, snapshot, or fork -- replacing fixed-percent auto-compaction across Claude Code, Codex, and MCP-capable hosts. Trigger on auto-compact timing or damage: "when should I compact", "is it safe to compact now or start a fresh session", "auto-compact fires too early/mid-task", "switching to an unrelated task but the window still has space", "context rot", "answers get worse the longer the session runs", "the agent forgot the plan or my decisions after it summarized", "add a layer on top that manages context without changing the agent", raising autoCompactWindow to give the policy room, or installing/tuning a cross-tool compaction policy or PreCompact hook -- even when "compaction" is never said but the problem is context-window pressure or post-summarization memory loss. Do NOT use to summarize a conversation, build RAG, write a summarization prompt (decides WHEN not HOW), or answer max-context-length trivia.
agent-skill-creator
IncludedCreate cross-platform agent skills from workflow descriptions. Activates when users ask to create an agent, automate a repetitive workflow, create a custom skill, or need advanced agent creation. Triggers on phrases like create agent for, automate workflow, create skill for, every day I have to, daily I need to, turn process into agent, need to automate, create a cross-platform skill, validate this skill, export this skill, migrate this skill. Supports single skills, multi-agent suites, transcript processing, template-based creation, interactive configuration, cross-platform export, and spec validation.
llm-wiki
IncludedUse when building or maintaining a persistent personal knowledge base (second brain) in Obsidian where an LLM incrementally ingests sources, updates entity/concept pages, maintains cross-references, and keeps a synthesis current. Triggers include "second brain", "Obsidian wiki", "personal knowledge management", "ingest this paper/article/book", "build a research wiki", "compound knowledge", "Memex", or whenever the user wants knowledge to accumulate across sessions instead of being re-derived by RAG on every query.
skill-master
IncludedAgent Skills authoring, evaluation, and optimization. Create, edit, validate, benchmark, and improve skills following the agentskills.io specification. Use when designing SKILL.md files, structuring skill folders (references, scripts, assets), ingesting external documentation into skills, running trigger evals, benchmarking skill quality, optimizing descriptions, or performing blind A/B comparisons. Keywords: agentskills.io, SKILL.md, skill authoring, eval, benchmark, trigger optimization.