drayline

Audits Claude Code repositories through a two-phase workflow. Phase 1 sweeps 14 hygiene categories and an optional 7-layer leak check, while Phase 2 executes [APPROVED] markers from the report file, useful for repo cleanup, pre-distribution sanity, and post-feature hygiene.

This skill produces structured XML session continuation documents for multi-session workflows, capturing all active work streams, decisions, files, conversation knowledge, and open items into an ingestion-optimized handoff file, and generates a closeout checklist.

Builds, validates, packages, tests, and optimizes deployment-ready Skill files from design specifications. It incorporates multiple quality gates and optimization tools, suitable for building, reviewing, revising, testing, or optimizing Skills.

Evaluates and scores Claude prompts and system prompts using a six-dimension Scorecard with anchored 1-5 rubrics. It triggers on phrases like 'review my prompt' or 'why isn't my prompt working'.

This skill offers 18 reasoning approaches across 6 categories for Claude prompts. It helps users retrieve, review, customize, or combine specific reasoning templates for their prompts.

Guides the selection of identity, reasoning, and output approaches for Claude prompts based on task characteristics. It triggers on specific queries like 'help me choose an approach' or 'recommend a prompt pattern', and also on symptom-phrased requests.

Detects seven structural anti-patterns in Claude Projects that cause unpredictable output, ignored instructions, and degraded quality. It diagnoses Monolith, Orphan File, Echo Chamber, Phantom Conversation, Kitchen Sink, Misaligned Hierarchy, and Blurred Layers.

Diagnoses and fixes Claude's behavioral issues in system prompts and Projects using tested countermeasure templates. It addresses common problems like Claude being too verbose, hedging, ignoring preferences, or adding unsolicited disclaimers.

This skill designs Claude Code prompts and CC environments, generating CLAUDE.md drafts, agent topologies, and remediation plans. It operates in five modes: DESIGN, EVOLVE, RESEARCH, TEMPLATE, and REMEDIATE.

This specialized prompt-building methodology for content and communications tasks offers 11 tested approaches (identity, reasoning, output). It's ideal for creating Claude prompts for content creation, editorial workflows, or communications tasks.

Specialized business strategy prompt methodology for Claude. It offers 11 tested approaches (3 identity, 4 reasoning, 4 output formats) for consulting, M&A, corporate strategy, and strategic planning tasks, ideal for prompts related to deal evaluation, due diligence, and strategic options assessment.

This specialized research and analysis prompt methodology for Claude provides 11 tested approaches (3 identity, 4 reasoning, 4 output). It's designed for data analysis, policy research, evidence review, and investigative research, aiding in quantitative data interpretation and evidence-based policy recommendations.

An independent re-derivation gate for proposed changes during autonomous execution, it evaluates changes (e.g., code diff, config edit, schema change) against an authority matrix and a 4-check protocol (invariant compliance, scope authorization, detection narrowness, regression risk). It returns structured JSON with pass/fail per check and an overall verdict (APPROVE / REQUEST_CHANGES).

Specialized agentic and context engineering prompt methodology for Claude, used in designing AI agents, including system prompts, tool definitions, context window architectures, failure mode planning, multi-agent coordination, and agent evaluation.

Analyzes Claude Project context budget with a two-pool architecture (knowledge files vs. threshold-exempt overhead). It offers Quick Diagnostic and Full Budget Audit modes, useful for optimizing token usage, checking context budget, or improving RAG retrieval quality.

Specialized software engineering prompt methodology for Claude. Use when building prompts for system design, code review, incident response, security analysis, API design, architecture decisions, RFCs, ADRs, runbooks, or technical leadership.

Audits and optimizes the five global Claude layers (User Preferences, Styles, Global Memory, Skills, MCP Connectors) using the Global Layer Scorecard (six dimensions, anchored 1-5 rubrics). It detects eight cross-layer failure modes and produces evolutionary recommendations (Promotion, Demotion, Codification, Skill Extraction).

Comprehensive audit of a user's entire Claude environment, combining Project and Global audits with a Cross-Layer Alignment Check and Evolutionary Recommendations. It produces a unified action plan.

Eight tested identity approaches for Claude prompts, each shaping depth, vocabulary, reasoning style, and what Claude treats as obvious versus requiring explanation. Use when the user wants a specific identity template — retrieving, reviewing, customizing, or building a role definition.

Evaluates if work-in-design is ready for autonomous execution from a chat design conversation (Claude Code, n8n, agent runtime). It uses a 7-condition gate covering spec stability, verification surface, invariants, pump-primer, work decomposition, rollback cost, and token budget headroom.

Rebalances context across Memory, Custom Instructions, knowledge files, and User Preferences in Claude Projects. It audits Memory for redundancy, staleness, and misplacement, prescribing optimization.

Selects the active profile from the triggering context by evaluating a rule set against various inputs (time, calendar, geofence, commands, signals). It returns the profile name for consuming Skills to apply, driven by a configuration file with deterministic precedence.

A schema-agnostic conversational profile builder that generates validated JSON configuration files for rootnode Skills. It conducts a progressive-depth interview using plain-language questions based on a JSON Schema (draft-07) to gather and validate information, then writes the resulting JSON.

Tests output format specifications for Claude prompts, used when a user requires a specific deliverable structure for retrieving, reviewing, customizing, or building output formatting. It provides 10 ready-to-use formats.

Generates a structured Project Brief, a comprehensive markdown document that extracts goals, architecture, and key decisions from a Claude Project. These briefs serve as uploadable context documents, providing immediate deep awareness of the source Project's purpose when added to another Project.

Audits and scores Claude Projects using a six-dimension Scorecard with anchored 1-5 rubrics and detects seven structural anti-patterns. When global layer info is provided, it adds cross-layer alignment findings.

Builds complete, optimized Claude prompts and Project scaffolds from task descriptions. Use when building, designing, creating, or structuring a Claude prompt or system prompt.