Claude Skills by Nova-Violet-Role

The AI_SLOP gate, the voice contract of every -dtd answer and, when the Adiutor is armed, of every answer, file, commit message and request body. Load when an answer reads generic, when the Adiutor closed a run with a slop finding, when an armed hook denied a Write, a commit or an answer for slop and the measures and the escape must be read, when a command's prose needs the ban list checked before it ships, when the bounds in ai-slop.dtd must be read or changed, or when a new record must not ...

The intake and decision gate as a reusable state machine. Load when a task should start with structured questions and a start, more, add gate, when designing a command that uses AskUserQuestion, or when a gate must be skipped safely in an autonomous run with every assumption listed.

Expert guidance for creating, writing, building, and refining Claude Code Skills. Use when working with SKILL.md files, authoring new skills, improving existing skills, or understanding skill structure and best practices. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

{{What it does}} Use when {{trigger conditions}}.

{{What it does}} Use when {{trigger conditions}}.

<required_reading> **Read these reference files NOW:** 1. references/recommended-structure.md 2. references/skill-structure.md 3. references/use-xml-tags.md </required_reading> <process>

<objective> Build a comprehensive execution skill that does real work in a specific domain. Domain expertise skills are full-featured build skills with exhaustive domain knowledge in references, complete workflows for the full lifecycle (build → debug → optimize → ship), and can be both invoked directly by users AND loaded by other skills (like create-plans) for domain knowledge. </objective> <critical_distinction> **Regular skill:** "Do one specific task" **Domain expertise skill:** "Do EVER...

<required_reading> **Read these reference files NOW:** 1. references/recommended-structure.md 2. references/skill-structure.md 3. references/core-principles.md 4. references/use-xml-tags.md </required_reading> <process>

<required_reading> **Read these reference files NOW:** 1. references/skill-structure.md </required_reading> <purpose> Audit checks structure. **Verify checks truth.** Skills contain claims about external things: APIs, CLI tools, frameworks, services. These change over time. This workflow checks if a skill's content is still accurate. </purpose> <process>

Expert guidance for creating, configuring, and using Claude Code hooks. Use when working with hooks, setting up event listeners, validating commands, automating workflows, adding notifications, or understanding hook types (PreToolUse, PostToolUse, Stop, SessionStart, UserPromptSubmit, etc). Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Create Model Context Protocol (MCP) servers that expose tools, resources, and prompts to Claude. Use when building custom integrations, APIs, data sources, or any server that Claude should interact with via the MCP protocol. Supports both TypeScript and Python implementations. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Create optimized prompts for Claude-to-Claude pipelines with research, planning, and execution stages. Use when building prompts that produce outputs for other prompts to consume, or when running multi-stage workflows (research -> plan -> implement). Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Create hierarchical project plans optimized for solo agentic development. Use when planning projects, phases, or tasks that Claude will execute. Produces Claude-executable plans with verification criteria, not enterprise documentation. Handles briefs, roadmaps, phase plans, and context handoffs. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

DTD-native: route a prompt to its schematic creator: ask the schematic, the semantic-schema families and the forms, then hand the purpose and every choice as known slots to create-prompt-<schematic>-dtd, which writes the file; this skill writes no prompt itself. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces

Expert guidance for creating Claude Code slash commands. Use when working with slash commands, creating custom commands, understanding command structure, or learning YAML configuration. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Expert guidance for creating, building, and using Claude Code subagents and the Task tool. Use when working with subagents, setting up agent configurations, understanding how agents work, or using the Task tool to launch specialized agents. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Deep analysis debugging mode for complex issues. Activates methodical investigation protocol with evidence gathering, hypothesis testing, and rigorous verification. Use when standard troubleshooting fails or when issues require systematic root cause analysis. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Audit one *-dtd artifact or the whole DTD corpus. Runs the rdc checker and dispatches the matching auditor agent (slash-command-auditor-dtd, skill-auditor-dtd, subagent-auditor-dtd, dtd-contract-auditor). Use before installing, after editing a shared subset, or when asked whether a -dtd file is sound.

The contract behind every *-dtd command, skill and agent. Load when writing, reading, installing or debugging a DTD-amplified artifact, when a DOCTYPE fails the rdc check, when PCDATA, CDATA, NDATA or NOTATION need to be applied to a prompt, or when a shared subset must be extended.

Route any request to the DTD-amplified commands by domain and type - read the request's semantic shape, match it against the domain table, and activate one family or chain several, with the ask gate and the cache bridge where the plan allows. Use when a request could need more than one command, when screenshots, research, caches or drifts are involved, or when a single no-gate command should suggest what runs next.

Measure whether a *-dtd command's answers conform to its declared grammar with the Adiutor as the instrument: run the command on a fixture argument, read the ledger line its Stop check wrote, then feed a deliberately broken answer through the same check in a scratch state directory and watch it fail. Use before shipping a new command, after changing a grammar, or when asked to prove a DOCTYPE is more than decoration.

Create a new *-dtd command, or convert an existing command into one, with a declared DOCTYPE, a trust boundary, a grammar map and laws, then prove it with the checker. Use when asked to write a DTD-amplified command, to add a Phantom Books style command, or to give an existing command a declared output grammar.

The forms a text may take inside a -dtd command and the guards between an untrusted text and a parser: shell heredocs in five variants, YAML block scalars in six, NestedText, JuliaMD, XML with CDATA, Markdown with the five GitHub callouts, JSON, TOML and the polyglots that are valid in more than one at once. Load when a creator asks which form an input or an output takes, when a command must render a heredoc, a block scalar or a callout without an injection path, when lib/form.mjs refused a t...

The IUPAC numerical multiplier prefixes (mono-, di-, tri-, icosa-, triaconta-, hecta-, kilia-) used as ordinals in file and directory names. Load when numbering a set of files with words instead of digits, when reading or writing a name like tri-extraction.md or docosa-appendix.md, when a composite prefix must be built for a number above twenty, or when a word-numbered directory has stopped sorting in the order its author intended. From 5.0.0 a record ordinal is the Greek cardinal (heis, duo,...

The Phantom Books corpus as a reference shelf for the nineteen book-derived commands (tetralemma, loci, babel, count-the-library, goetia, clean-unclean, eleusis, voluspa, havamal, atharvan, sutra, wu-wei, water, witnesses, four-branches, redaction, sapiential, catalog, formula). Load when running one of them and the book's actual structure matters, when adding a new book-derived command, or when asked which book a command draws on.

The numbered, append-only field discipline for any file one session writes and a later session parses: handoffs, todo lists, plans, indexes, TSV logs. Load when declaring a RECORD.* entity, when adding a column to an existing record, when a reader finds more or fewer columns than expected, or when a file format must survive across versions.

The nine RoT MoE lenses and the MoE engine as declared grammar. Load when running any /rot-*-dtd command, when a lens's parameters (lambda, mu, entropy band, gauge band, bound), its experts or its interceptors are needed, when the TIER 1 stems or a weight profile must be read, when the PRISM gauge must be computed, when two lenses must be composed into a hybrid by the law, when the live router marker must be read, or when a new lens-derived command is being written.

Run saved prompts in isolated sub-agent contexts, single, sequential or parallel. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Set up and configure Geoffrey Huntley's original Ralph Wiggum autonomous coding loop in any directory with proper structure, prompts, and backpressure. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

The AI_SLOP gate, the voice contract of every -dtd answer and, when the Adiutor is armed, of every answer, file, commit message and request body. Load when an answer reads generic, when the Adiutor closed a run with a slop finding, when an armed hook denied a Write, a commit or an answer for slop and the measures and the escape must be read, when a command's prose needs the ban list checked before it ships, when the bounds in ai-slop.dtd must be read or changed, or when a new record must not ...

The intake and decision gate as a reusable state machine. Load when a task should start with structured questions and a start, more, add gate, when designing a command that uses AskUserQuestion, or when a gate must be skipped safely in an autonomous run with every assumption listed.

Expert guidance for creating, writing, building, and refining Claude Code Skills. Use when working with SKILL.md files, authoring new skills, improving existing skills, or understanding skill structure and best practices. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

{{What it does}} Use when {{trigger conditions}}.

{{What it does}} Use when {{trigger conditions}}.

<required_reading> **Read these reference files NOW:** 1. references/recommended-structure.md 2. references/skill-structure.md 3. references/use-xml-tags.md </required_reading> <process>

<objective> Build a comprehensive execution skill that does real work in a specific domain. Domain expertise skills are full-featured build skills with exhaustive domain knowledge in references, complete workflows for the full lifecycle (build → debug → optimize → ship), and can be both invoked directly by users AND loaded by other skills (like create-plans) for domain knowledge. </objective> <critical_distinction> **Regular skill:** "Do one specific task" **Domain expertise skill:** "Do EVER...

<required_reading> **Read these reference files NOW:** 1. references/recommended-structure.md 2. references/skill-structure.md 3. references/core-principles.md 4. references/use-xml-tags.md </required_reading> <process>

<required_reading> **Read these reference files NOW:** 1. references/skill-structure.md </required_reading> <purpose> Audit checks structure. **Verify checks truth.** Skills contain claims about external things: APIs, CLI tools, frameworks, services. These change over time. This workflow checks if a skill's content is still accurate. </purpose> <process>

Expert guidance for creating, configuring, and using Claude Code hooks. Use when working with hooks, setting up event listeners, validating commands, automating workflows, adding notifications, or understanding hook types (PreToolUse, PostToolUse, Stop, SessionStart, UserPromptSubmit, etc). Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Create Model Context Protocol (MCP) servers that expose tools, resources, and prompts to Claude. Use when building custom integrations, APIs, data sources, or any server that Claude should interact with via the MCP protocol. Supports both TypeScript and Python implementations. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Create optimized prompts for Claude-to-Claude pipelines with research, planning, and execution stages. Use when building prompts that produce outputs for other prompts to consume, or when running multi-stage workflows (research -> plan -> implement). Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Create hierarchical project plans optimized for solo agentic development. Use when planning projects, phases, or tasks that Claude will execute. Produces Claude-executable plans with verification criteria, not enterprise documentation. Handles briefs, roadmaps, phase plans, and context handoffs. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

DTD-native: route a prompt to its schematic creator: ask the schematic, the semantic-schema families and the forms, then hand the purpose and every choice as known slots to create-prompt-<schematic>-dtd, which writes the file; this skill writes no prompt itself. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces

Expert guidance for creating Claude Code slash commands. Use when working with slash commands, creating custom commands, understanding command structure, or learning YAML configuration. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Expert guidance for creating, building, and using Claude Code subagents and the Task tool. Use when working with subagents, setting up agent configurations, understanding how agents work, or using the Task tool to launch specialized agents. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Deep analysis debugging mode for complex issues. Activates methodical investigation protocol with evidence gathering, hypothesis testing, and rigorous verification. Use when standard troubleshooting fails or when issues require systematic root cause analysis. Carries its own DOCTYPE: a declared output grammar, a trust boundary and laws the checker enforces.

Audit one *-dtd artifact or the whole DTD corpus. Runs the rdc checker and dispatches the matching auditor agent (slash-command-auditor-dtd, skill-auditor-dtd, subagent-auditor-dtd, dtd-contract-auditor). Use before installing, after editing a shared subset, or when asked whether a -dtd file is sound.

The contract behind every *-dtd command, skill and agent. Load when writing, reading, installing or debugging a DTD-amplified artifact, when a DOCTYPE fails the rdc check, when PCDATA, CDATA, NDATA or NOTATION need to be applied to a prompt, or when a shared subset must be extended.

Route any request to the DTD-amplified commands by domain and type - read the request's semantic shape, match it against the domain table, and activate one family or chain several, with the ask gate and the cache bridge where the plan allows. Use when a request could need more than one command, when screenshots, research, caches or drifts are involved, or when a single no-gate command should suggest what runs next.

Measure whether a *-dtd command's answers conform to its declared grammar with the Adiutor as the instrument: run the command on a fixture argument, read the ledger line its Stop check wrote, then feed a deliberately broken answer through the same check in a scratch state directory and watch it fail. Use before shipping a new command, after changing a grammar, or when asked to prove a DOCTYPE is more than decoration.