Synthesize the current conversation context into a flow-next spec at `.flow/specs/<spec-id>.md` via `flowctl spec create + spec set-plan` — agent-native, source-tagged, with mandatory read-back before write. Triggers on /flow-next:capture, "capture spec", "lock down what we discussed", "make a spec from this conversation", "convert conversation to spec". Optional `mode:autofix` token runs without questions and requires `--yes` to commit. Optional `--rewrite <spec-id>` overwrites an existing spec; `--from-compacted-ok` overrides the compaction-detection refusal; `--override-strategy` proceeds despite a contradiction with an active STRATEGY.md track (and prompts to record the override as a decision).

Configure, extend, or contribute to Hermes Agent.

This skill should be used when users want to build LLM-powered applications using LangChain. It provides patterns for initializing any LLM provider (OpenAI, Anthropic, Google, xAI), building agent loops with tools, and implementing structured output. Use this skill when users ask to create chatbots, AI agents, or applications that need LLM integration with tool calling or structured responses.

Optimize LLM-facing content: documentation c7score, llms.txt generation, Claude Code skill optimization, XML tag structuring for prompts, CLAUDE.md auditing, and LLM parameter tuning. Use when optimizing docs for AI assistants, creating llms.txt, improving c7score, optimizing skills for token efficiency, applying 500-line rule, implementing progressive disclosure, designing XML tag structures for prompts, auditing CLAUDE.md files, tuning LLM temperature and token settings, or compressing markdown for context efficiency.

Land a subagent worktree branch onto an integration branch with preflight + recorded rollback. Triggers: operator (or `next_actions`) surfaces verb `merge_orchestrate` via Exarchos MCP. Local git operation — NOT remote PR merging (that is `merge_pr`).

Comprehensive skill for project planning and prompt engineering. Covers hierarchical plans (briefs, roadmaps, phases), Claude-to-Claude meta-prompts, and multi-stage workflows. Use when: planning, prompt creation, agentic pipeline work, project roadmap, meta-prompts, research to implement workflow.

Craft, update, or review LLM prompts from first principles. Use when creating new prompts, updating existing ones, reviewing prompt structure, or diagnosing a failing prompt. Ensures prompts define WHAT and WHY, not HOW. Triggers: write a prompt, edit a prompt, review a prompt, improve a prompt, diagnose prompt failure, fix failing prompt, system prompt, skill, agent.

Analyzes and optimizes prompts using BP-001~008 patterns and 3-step flow (detect, optimize, balance). Use when "optimize this prompt", "review prompt quality", "analyze prompt issues", or creating/reviewing rashomon skill content.

Create new Skills for Claude Code including YAML frontmatter, descriptions, instructions, and structure. Use when creating, building, or designing skills, or when asked about skill creation, structure, best practices, or debugging skills that don't activate properly.

Twilio Conversation Intelligence development guide. Use when building real-time or post-call conversation analysis, language operator pipelines, sentiment analysis, agent assist, cross-channel analytics, or querying aggregated conversation insights (sentiment trends, escalation rates, dashboards).

Provides knowledge about acpx CLI for agent-to-agent communication. Use when user asks about acpx commands, ACP protocol, agent sessions, prompt queueing, or scriptable agent workflows.

Vector database for embeddings and semantic search at the edge. Load when building RAG pipelines, implementing semantic search, storing embeddings from Workers AI, finding similar documents, building recommendation systems, or querying vectors by similarity.

This skill should be used when designing context management, implementing tiered fidelity, reducing token waste, applying Four Laws patterns, creating "NOT PASSED" sections, optimizing agent context, or debugging context-related issues. Provides SOTA patterns for context-efficient multi-agent systems achieving 60-80% token reduction.

Backfill `<!-- EXPLORER_HEADER ... -->` blocks into existing `docs/NN-*.md` and `docs/components/**/*.md` files that predate v3.14.0. The header (5–10 lines after the H1) surfaces key concepts, technologies, components, scope, and related ADRs in the first 60 lines so the architecture-explorer agent classifies the file accurately. Invoke when a doc was created before v3.14.0, when bulk-backfilling a legacy project, or whenever the explorer is misclassifying files.

This skill should be used when the user asks to "create a hook", "add a hook", "write a hook", or mentions Claude Code hooks. Also suggest this skill when the user asks to "automatically do X" or "run X before/after Y" as these are good candidates for hooks.

CLAUDE.md file generation and optimization for Claude Code projects. Capabilities: initialize project instructions, analyze codebase context, optimize existing CLAUDE.md, apply Anthropic best practices, reduce token usage, improve effectiveness. Actions: init, create, optimize, enhance CLAUDE.md files. Keywords: CLAUDE.md, project instructions, Claude Code setup, project context, codebase analysis, Anthropic best practices, token optimization, project configuration, AI instructions, coding guidelines, project rules, workspace setup. Use when: initializing CLAUDE.md for new projects, optimizing existing project instructions, setting up Claude Code for a codebase, improving AI coding guidelines.

Guide for Claude Code dynamic workflows — JavaScript scripts that orchestrate many subagents in one background run. Use when a task needs more agents than one conversation can coordinate, codifying repeatable orchestration, running codebase-wide sweeps or large migrations, authoring or reading a workflow script, or deciding between a workflow, a subagent, and a skill.

Use when working on Claude Code plugins (creating, modifying, testing, releasing, or maintaining) - provides streamlined workflows, patterns, and examples for the complete plugin lifecycle

Multi-model consensus scoring for content ideas. Scores the same idea with Claude, GPT-4o, Gemini, and Grok in parallel, then aggregates for a balanced verdict. Reduces single-model bias and improves viral predictions.

Comprehensive expertise for working with Microsoft's GenAIScript framework - a JavaScript/TypeScript-based system for building automatable LLM prompts and AI workflows. Use when creating, debugging, or optimizing GenAIScript scripts, implementing prompts-as-code, working with tools and agents, processing files (PDF, CSV, DOCX), defining schemas, or building AI automation workflows.

Use PROACTIVELY when working on multiple branches simultaneously or creating parallel Claude Code sessions. Automates git worktree creation with prerequisite checking, development environment initialization, and safe cleanup. Supports single worktree, batch creation, and worktree removal. Not for non-git projects or branch management without worktrees.

Prompt engineering guidance for Grok (xAI) model. Use when crafting prompts for Grok to leverage its conversational style, real-time knowledge access, and less-constrained responses.

Automated LLM-driven hypothesis generation and testing on tabular datasets. Use when you want to systematically explore hypotheses about patterns in empirical data (e.g., deception detection, content analysis). Combines literature insights with data-driven hypothesis testing. For manual hypothesis formulation use hypothesis-generation; for creative ideation use scientific-brainstorming.

Run high-stakes questions, ideas, and decisions through a council of 5 AI advisors who analyze independently, peer-review each other anonymously, and synthesize a final verdict with artifacts. Mandatory triggers: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. Strong triggers when paired with a real decision or tradeoff: 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do not use for factual lookups, simple yes/no questions, summarization, or low-stakes choices.

Integration patterns for Mapbox MCP DevKit Server in AI coding assistants. Covers setup, style management, token management, validation workflows, and documentation access through MCP. Use when building Mapbox applications with AI coding assistance.

Senior MLOps Engineer with 8+ years ML systems experience. Use for model serving & inference infrastructure, AI/ML pipelines, training-data pipelines, model deployment & monitoring, and AI cost optimization at the infrastructure level. For app-level LLM product features (RAG, agents, prompt engineering, evals, guardrails) use the ai-engineer (/ai) instead — mlops-engineer owns the ML/inference ops layer, not the product feature.

Generate and improve prompts using best practices for OpenAI GPT-5 and other LLMs. Apply advanced techniques like chain-of-thought, few-shot prompting, and progressive disclosure.

Converts Agent Alchemy plugins into formats compatible with other AI coding platforms

Multi-agent review resolution with trust-but-verify methodology. Accepts a PR number/URL or a local code-review report file. Extracts review comments, verifies each with parallel agents using confidence scoring (0-100), filters false positives, presents a triage report for approval, then applies fixes and replies. Use when user runs /resolve-review, mentions "resolve PR comments", "address review feedback", "fix PR suggestions", "handle review comments", or "resolve review findings".

Use after completing work sessions to analyze agent behavior patterns, prepare session handoffs for continuity, document completed work, identify blockers, or preserve context for the next session.

Focused Signals scout for PostHog projects using AI observability. Rotates through a set of lenses — cost, latency, errors, volume, eval performance, eval/enrichment config, clusters, and tool usage — watching each for trends and spikes sliced by the dimensions it discovers over time. Leans on the sandbox's bundled `exploring-llm-*` deep-dive skills for the actual queries. Emits findings only when they clear the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — no dependencies on other scouts. Picked uniformly at random by the coordinator alongside `signals-scout-general` and other specialists.

Audit a skill against Anthropic's production best practices (9-point scorecard). Use after creating or improving any skill via /skill-creator. Also use when user says 'audit skill', 'check skill quality', 'review my skill', or wants to validate a skill before distribution. NOT for project-level audits (/audit) or harness health (/harness-audit).

Route tasks to agents using Twilio TaskRouter. Covers Workers, Task Queues, Workflows, Reservations, skills-based routing, and common gotchas (hyphen attributes, HAS operator, reservation cascade). Use this skill for any multi-agent contact center, support queue, or AI agent escalation routing.

Deep multi-platform intelligence search across ALL AnySite MCP sources (LinkedIn, Twitter, Instagram, Reddit, Y Combinator, web). Auto-detects search type (person, company, or topic) and performs CASCADING research - person searches include their company analysis, company searches include leadership profiles, topic searches aggregate facts from all platforms. Use when users ask to "find", "search", "research", "investigate", or need comprehensive intelligence on any subject. Produces detailed reports with cross-validated findings and source links.

Answer questions about the Mastra AI framework and help build agents, workflows, tools, memory, and RAG features. Use when developers: (1) ask about Mastra APIs like `Mastra`, `Agent`, `createWorkflow`, `createTool`, `createStep`, `Memory`, or run the `mastra` CLI; (2) wire up agents, workflows, or step-based orchestration; (3) integrate `@mastra/*` packages (`@mastra/core`, `@mastra/memory`, `@mastra/rag`, `@mastra/evals`, `@mastra/deployer`, `@mastra/server`, `@mastra/pg`, `@mastra/libsql`, `@mastra/mcp`); (4) debug TypeScript errors from Mastra APIs or model-router strings. Triggers on: "Mastra", "@mastra/core", "mastra agent", "mastra workflow", "createWorkflow", "createTool", "createStep", "mastra memory", "semantic recall", "mastra rag", "mastra deployer", "mastra evals", "mastra studio", "provider/model".

Use this skill when an agent CLI's behavior should be versioned data rather than code strings — managing prompts, policies, and output schemas — or when standing up an autonomous loop (scheduled/event-driven runs with guarded apply) and its maintain/release rails. Triggers on "externalize the prompt", "version the agent's policy", "the agent's output schema", "make this CLI run on a schedule", "autonomous bot", "agent decision contract", "set up the maintainer loop", "release the CLI". Distilled from openclaw/clawsweeper.

Deploys swarms of sub-agents for massive parallel data processing tasks. Unlike agent-army (which is for code changes), this is for DATA tasks -- processing 1000 documents, analyzing datasets, bulk content generation. Configurable swarm size, task distribution, result aggregation, progress tracking, and error recovery.

Designs and deploys custom agent teams for specific business workflows. Interactive discovery of business processes, then generates complete team configurations with specialized agent roles, tool access, communication protocols, and handoff rules.

Framework for self-sustaining AI agent teams with self-service work queues, role ownership, continuous discovery, and proactive operation via heartbeat loops.

Create and manage Claude Code skills following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns, file paths, content patterns), enforcement levels (block, suggest, warn), hook mechanisms (UserPromptSubmit, PreToolUse), session tracking, and the 500-line rule.

Master Anthropic's prompt engineering techniques to generate new prompts or improve existing ones using best practices for Claude AI models.

# Arc Security - Agent Trust Protocol

Show recent Claude Code changes personalized to user's setup. Trigger when user asks about changelog, "what's new", "cc changelog", "claude code updates", "recent changes", or wants to know what changed in Claude Code.

This skill should be used when the user asks to "set up memory", "configure CLAUDE.md", "add rules", "create project rules", "organize instructions", or mentions Claude Code memory, CLAUDE.md files, .claude/rules, rules, or project instructions. Also use when the user wants to persist preferences, add coding conventions, scope instructions to specific files, or understand when to use rules vs hooks or other memory mechanisms.

This skill should be used when the user asks to "create a statusline", "customize the status line", "add a custom prompt", or mentions Claude Code statusline. Also suggest when the user wants to display git branch, context usage, model name, or session costs at the bottom of Claude Code.

Multi-agent code review using the Pragmatic Quality framework. Orchestrates parallel review agents (CLAUDE.md compliance, bug scanning, git history, security) with independent confidence scoring to produce high-signal findings. Use when user runs /code-review, /review:code-review, requests a "code review", "review my changes", "PR review", or mentions "review diff", "review branch".

Iterative non-code discussion between the local agent and Codex CLI on any open-ended topic: diet, fitness, writing, decisions, strategy, study plans, life choices, brainstorming. Orchestrates an automatic back-and-forth debate where both agents critique, propose alternatives, and iterate on the user's idea until reaching consensus. Codex CLI runs READ-ONLY, forms its own opinions, and normally does not navigate the filesystem unless the user provides file paths. Use when the user says discuss with codex, iterate with codex, consult codex, debate with codex, ask codex for a second opinion, get codex's take, or brainstorm with codex, including pasting or describing a plan, draft, idea, decision, or proposal and wanting a critical iterative review. Does NOT trigger on code review, plan-mode review of implementation plans, architecture discussions, or any technical software-engineering analysis; use codex-review for those.

This skill should be used when the user asks to "run codex", "use codex CLI", "delegate to codex", "codex resume", or "continue with codex". Executes tasks via OpenAI Codex CLI with model selection, reasoning effort configuration, and session management.

Expert guidance for creating Claude Code slash commands. Use when working with slash commands, creating custom commands, understanding command structure, or learning YAML configuration.

Comprehensive CrewAI framework guide for building collaborative AI agent teams and structured workflows. Use when developing multi-agent systems with CrewAI, creating autonomous AI crews, orchestrating flows, implementing agents with roles and tools, or building production-ready AI automation. Essential for developers building intelligent agent systems, task automation, and complex AI workflows.

This skill should be used when the user asks to "deep research", "comprehensive research on", "thorough investigation of", "research report on", "deep dive into", "literature review on", or needs Gemini Deep Research for web-grounded multi-source synthesis beyond what Google Scholar and Consensus provide.

Develops, optimizes, and validates Claude Code plugins, skills (SKILL.md), commands, agents, hooks (hooks.json), and scripts. Ensures consistency with official best practices. Activates when creating, editing, or reviewing files in plugins/ directory, .claude-plugin/, plugin.json, permissions.json, or marketplace.json. Covers YAML frontmatter, element structure, cross-references, naming conventions, and plugin manifest validation. NOT for application development (use domain-specific skills).

Distill a session-tethered working context into a self-contained portable handoff a fresh zero-memory agent can execute from. Closes deictic, grounding, relevance, provenance, and compression gaps, then emits a prose channel plus a schema-versioned TaskStateBlock. Type: (ContextTethered, AI, DISTILL, WorkingContext) → PortableHandoff. Alias: Diylisis(διύλισις).

Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Graceful rate limit handling with Ollama fallback. Notifies on rate limits, offers local model switch with confirmation for code tasks.

Maintains a persistent, LLM-managed personal wiki. Activate when the user wants to ingest a source ("ingest this", "add to wiki", "process this article", "file this"), query their knowledge base ("what does my wiki say", "search my wiki", "answer from my notes"), lint the wiki ("lint wiki", "find orphan pages", "check wiki health", "clean up wiki"), or bootstrap a new wiki ("create wiki", "set up knowledge base", "initialize llm-wiki"). Handles single-source ingest workflows, index-first query synthesis, and structural health checks.

Route AI coding queries to local LLMs in air-gapped networks. Integrates Serena MCP for semantic code understanding. Use when working offline, with local models (Ollama, LM Studio, Jan, OpenWebUI), or in secure/closed environments. Triggers on local LLM, Ollama, LM Studio, Jan, air-gapped, offline AI, Serena, local inference, closed network, model routing, defense network, secure coding.

Connect to any MCP (Model Context Protocol) server (Including Robinhood and Base) and use its tools, resources, and prompts

Run multiple AI models in parallel for 3-5x speedup with ENFORCED performance statistics tracking. Use when validating with Grok, Gemini, GPT-5, DeepSeek, or Claudish proxy for code review, consensus analysis, or multi-expert validation. NEW in v3.1.0 - SubagentStop hook enforces statistics collection, MANDATORY checklist prevents incomplete reviews, timing instrumentation examples. Includes dynamic model discovery via `claudish --top-models` and `claudish --free`, session-based workspaces, and Pattern 7-8 for tracking model performance. Trigger keywords - "grok", "gemini", "gpt-5", "deepseek", "claudish", "multiple models", "parallel review", "external AI", "consensus", "multi-model", "model performance", "statistics", "free models".

Use this skill when launching Task agents, batch processing multiple files, delegating work to sub-agents, or running parallel operations. Provides XML prompt structure, agent type selection, and coordination patterns.

Act as a thought partner overseeing multiple agent sessions and controllers. Use when asked to check on agents, give a status update, oversee work, coordinate between workers, act as a controller for the day, or help prioritize across workstreams. Triggers: "oversee", "check on agents", "status update", "be my controller", "what are my workers doing", "coordinate", "thought partner", "what should I focus on".

Validate, test, and distribute Claude Code plugins and marketplaces. Use when developing plugins, debugging validation errors, or preparing for distribution.

Transform prompts into structured TCRO format with phase-specific clarification. Automatically invoked by /ai-eng/research, /ai-eng/plan, /ai-eng/work, and /ai-eng/specify commands. Use when refining vague prompts, structuring requirements, or enhancing user input quality before execution.

Summarises the delta between a tool's latest release and the last summary the user saw. Use when the user asks about what's new, the latest release, release notes, or the changelog of one of the supported tools: Claude Code, OpenCode, llama-swap, or llama.cpp.

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.

Evaluates which SaaS tools can be replaced with AI agents. Takes a list of current SaaS subscriptions with costs, assesses replacement feasibility, estimates build vs buy economics, identifies Claude+MCP alternatives, and generates a comprehensive replacement plan with priority matrix, ROI analysis, implementation timeline, and risk assessment.

Security analysis and threat modeling prompt templates for STRIDE analysis, code review, OWASP compliance, and vulnerability assessment. Use for security planning, pre-deployment reviews, and ongoing threat assessment. Triggers include "STRIDE", "threat model", "security review", "code review", "OWASP", "payment security", "security analysis", "vulnerability assessment".

Create and manage Claude Code skills following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns, file paths, content patterns), enforcement levels (block, suggest, warn), hook mechanisms (UserPromptSubmit, PreToolUse), session tracking, and the 500-line rule.

Token optimization best practices for cost-effective Claude Code usage. Automatically applies efficient file reading, command execution, and output handling strategies. Includes model selection guidance (Opus for learning, Sonnet for development/debugging). Prefers bash commands over reading files.

Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.

Add agent-friendly --json NDJSON output to Python CLI scripts, or scaffold a complete cli_utils package for a project. Use this skill when the user wants to make scripts machine-readable for AI agents, add --json flags, convert print statements to structured JSON, build a CLI helper library, create an open-source CLI-for-agents package, add structured logging, or make CLI output machine-readable. Also use when the user mentions NDJSON, structured CLI output, agent-friendly CLI, non-interactive scripts, or JSON I/O for automation.

Implementation + audit loop using parallel agent teams with structured simplify, harden, and document passes. Spawns implementation agents to do the work, then audit agents to find complexity, security gaps, and spec deviations, then loops until code compiles cleanly, all tests pass, and auditors find zero issues or the loop cap is reached. Use when: implementing features from a spec or plan, hardening existing code, fixing a batch of issues, or any multi-file task that benefits from a build-verify-fix cycle.

Create an AI Evals Pack (eval PRD, test set, rubric, judge plan, results + iteration loop). Use for LLM evaluation, benchmarks, rubrics, error analysis/open coding, and ship/no-ship quality gates for AI features.

Competitive intelligence gathering using anysite MCP server across LinkedIn, social media, Y Combinator, and the web. Track competitor activities, analyze hiring patterns, monitor content strategies, benchmark market positioning, research startup competitors, and gather strategic intelligence. Supports LinkedIn (companies, employees, posts), Instagram, Twitter/X, Reddit, YouTube, Y Combinator, and web scraping. Use when users need to analyze competitors, track competitive movements, research market positioning, monitor hiring velocity, or gather strategic market intelligence.

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

Activation-aware weight quantization for 4-bit LLM compression with 3x speedup and minimal accuracy loss. Use when deploying large models (7B-70B) on limited GPU memory, when you need faster inference than GPTQ with better accuracy preservation, or for instruction-tuned and multimodal models. MLSys 2024 Best Paper Award winner.

Authors and updates customization overrides for installed BMad skills. Use when the user says 'customize bmad', 'override a skill', 'change agent behavior', or 'customize a workflow'.

Universal ChromaDB integration patterns for semantic search, persistent storage, and pattern matching across all agent types. Use when agents need to store/search large datasets, build knowledge bases, perform semantic analysis, or maintain persistent memory across sessions.

Browser automation via Chrome extension and DevTools MCP. Use when controlling a logged-in Chrome browser, automating workflows, filling forms, extracting data, debugging web apps, reading console logs, or running scheduled browser tasks. Use for browser testing, form automation, data extraction, web scraping, authenticated browsing.

This skill should be used when the user asks to "create a plugin", "build a plugin", "write a plugin", or wants to bundle agents, hooks, commands, skills, or MCP servers into a distributable Claude Code plugin.

Capture learnings from the current coding session and update AGENTS.md. Use when the user asks to close the loop, run session-commit, record best practices, or update agent instructions based on recent work.

Generate interactive AI transformation context-builder prompts for consulting clients. Use when creating structured discovery session prompts that guide a company through context gathering about their business, pain points, tech stack, and AI opportunities. Produces a resumable, multi-section prompt with Express/Deep Dive modes.

Strategies for managing LLM context windows effectively in AI agents. Use when building agents that handle long conversations, multi-step tasks, tool orchestration, or need to maintain coherence across extended interactions.

Global hooks for logging Claude Code conversation events to markdown files. Tracks prompts, tool usage, and responses across all sessions. Useful for debugging, auditing, and providing conversation context to Claude.

Generate self-contained, agent-executable Execution Plans (ExecPlans) from PRD files. Use this skill whenever the user wants to "create a plan", "generate an execution plan", "turn a PRD into a plan", "write an ExecPlan", "create an implementation plan from a PRD", "break down a PRD into steps", "make a plan from requirements", "convert PRD to plan", or mentions "ExecPlan", "execution plan", or "implementation plan" in the context of producing one from a requirements document. Also triggers when the user has a PRD.md file and wants actionable next steps, says "plan the implementation", "create milestones from this PRD", or "how do I implement this PRD". This skill produces plans that a stateless coding agent or human novice can follow from top to bottom to deliver working, observable results.

Provides AI and machine learning techniques for CTF challenges. Use when attacking ML models, crafting adversarial examples, performing model extraction, prompt injection, membership inference, training data poisoning, fine-tuning manipulation, neural network analysis, LoRA adapter exploitation, LLM jailbreaking, or solving AI-related puzzles.

This skill should be used when the user asks to evaluate a completed team spawn session, capture session learnings, or review how a spawn went. Use this skill when the user says "evaluate spawn", "evaluate-spawn", "how did the spawn go", "capture learnings", "review spawn session", "spawn retrospective", or at the soft prompt after a team completes its work. This is a voluntary post-spawn evaluation with profile-based questions tailored to all spawn types (build, think, create). Build profile asks up to 4 questions (3 explicit + 1 conditional gate bypass); Think and Create profiles ask up to 3 questions (2 explicit + 1 gate bypass). Hard cap includes auto-derived scenario coverage for Build.

Comprehensive codebase assessment for agent and production readiness. Scans 8 pillars (48 criteria), verifies commands work, checks GitHub settings. Reports everything, fixes agent readiness only. Triggers on /flow-next:prime.

Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Interact with a running Godot game via MCP — launch, screenshot, click, inspect scene tree, get/set properties. Powered by godot-mcp + GodotMCPBridge autoload.

Generate hierarchical AGENTS.md structure for codebases to optimize AI agent token usage. Use when creating AGENTS.md files, documenting codebase structure, setting up agent guidance, organizing project documentation for AI tools, implementing JIT indexing, or working with monorepos that need lightweight root guidance with detailed sub-folder documentation. Covers repository analysis, root AGENTS.md generation, sub-folder AGENTS.md creation, and token-efficient documentation patterns.

Create well-formed Linear issues on command for any project. Accepts a GitHub URL or local disk path as project context, explores the codebase and documentation adaptively, then creates a structured issue using DEV-007 templates (Bug, Feature, Tech-Debt, Idée, Hotfix) via the Linear MCP.

Root cause analysis on production LLM traces. Diagnoses why an LLM application is failing — works from eval judge verdicts, runtime errors, or structural anomalies depending on what signals are present. Walks the span tree from symptom to root cause. Use when user says "what's wrong with my app", "why is my eval failing", "analyze errors", "root cause analysis", "diagnose failures", or wants to understand production failure patterns.

Comprehensive LLM security testing prompts for bias detection, data leakage, alignment testing, and adversarial prompt resistance.

Converts baseball and fantasy-baseball jargon into plain English for a user with zero baseball knowledge. Wraps every user-facing sentence produced by the MLB agent team (morning briefs, trade recommendations, waiver calls, chat summaries). Detects jargon terms, attaches an inline parenthetical plain-English gloss on first mention in a document, enforces the action-verb ladder (START / SIT / ADD / DROP / BID $X / ACCEPT / COUNTER / REJECT), and rejects assumed-knowledge phrases like "hot streak" or "positive matchup." Use when asked to translate for beginner, explain in plain English, translate this, write without jargon, make it beginner-friendly, or produce any user-facing MLB output for K L D'Souza's Fantasy Baseball 2K25 team.

Computes FAAB (Free Agent Acquisition Budget) recommended and maximum bids for Yahoo fantasy baseball waiver targets. Implements the baseball-specific layering of the faab-bid-framework (positional_need_fit, role_certainty, urgency, season_pace, league-inflation calibration) and DELEGATES the game-theoretic primitives -- first-price shading and winner's-curse haircut -- to the sibling skills `auction-first-price-shading` and `auction-winners-curse-haircut`. Produces a recommended bid, a hard ceiling, a rationale with the full delegation chain, and guardrail flags. Use when the user asks "how much should I bid on X", mentions FAAB bid, waiver bid amount, blind bid, Yahoo waiver claim sizing, or when mlb-waiver-analyst needs a bid amount for an identified target.