Generate optimized handoff prompts for delegating work to another LLM agent. Use when handing work to GPT-5.x/Codex, Claude 4.x, Gemini 3.x, or Grok 4.x, either as a shared-workspace sub-task handoff or a fresh-context handoff for a new session or model. Triggers on requests like "create a handoff prompt", "delegate this task to another agent", "hand this off", or "prepare context for another agent".

Operate the joelclaw event bus, gateway, observability, and agent loop infrastructure via the joelclaw CLI. Use for: sending events, checking runs, starting/monitoring/cancelling agent loops, debugging failed runs, checking health, restarting the worker, inspecting step traces, managing subscriptions, gateway operations, OTEL queries, semantic recall. Triggers: 'joelclaw', 'send an event', 'check inngest', 'start a loop', 'loop status', 'why did this fail', 'debug run', 'check worker', 'restart worker', 'runs', 'what failed', 'subscribe', 'gateway', 'otel', 'recall', or any Inngest/event-bus/agent-loop task.

Coordinate multiple agents in parallel or sequential workflows. Use when running agents simultaneously, delegating to sub-agents, switching between specialized agents, or managing agent selection. Trigger keywords - "parallel agents", "sequential workflow", "delegate", "multi-agent", "sub-agent", "agent switching", "task decomposition".

Coordinate multiple agents in parallel or sequential workflows. Use when running agents simultaneously, delegating to sub-agents, switching between specialized agents, or managing agent selection. Trigger keywords - "parallel agents", "sequential workflow", "delegate", "multi-agent", "sub-agent", "agent switching", "task decomposition".

Screen incoming NDAs and classify them as GREEN (standard), YELLOW (needs review), or RED (significant issues). Use when a new NDA comes in from sales or business development, when assessing NDA risk level, or when deciding whether an NDA needs full counsel review.

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. Originally from OpenAI's curated skills catalog.

Detailed implementation strategies for the four-tier orchestrator agent architecture including sub-agent definitions and group coordination protocols.

Load when an agent needs to manage, install, update, remove, or validate OpenViking skills via the `ov skills` CLI. Trigger on explicit user requests about skill management, when the user mentions `ov skills`, `install skill`, `update skill`, `delete skill`, `validate skill`, or when an agent needs to discover what skills are available on the OpenViking server.

Orchestrates a full code review for the current branch or a GitHub PR. Detects context, delegates to code-reviewer agent, displays the report, then optionally posts to the PR as line comments or a single comment via gh-pr-review. Use when the user asks for a code review, PR review, or audit.

Resolve and load any agent capability across the whole fleet from one place. Given an intent, finds the matching skill, command, or agent across Claude, Codex, Gemini, .agents, a-i--skills, OpenCode, and OpenClaw — then loads it through a single portal. Use for cross-agent skill discovery, "which agent has a skill for X", capability routing, or "find a skill that does Y" when capabilities are spread across multiple agent ecosystems.

Resources for conversational interviews to create learner profiles. Used by project-interviewer agent during /init.

A skill for improving prompts by applying general LLM/agent best practices. When the user provides a prompt, this skill outputs an improved version, identifies missing information, and provides specific improvement points. Use when the user asks to "improve this prompt", "review this prompt", or "make this prompt better".

Fan-out search across all memory sources when context is unclear or vaguely referenced. Triggers on: 'from earlier', 'remember when', 'what we discussed', 'that thing with', 'the conversation about', 'did we ever', 'what happened with', 'you mentioned', 'we talked about', 'earlier today', 'last session', 'the other day', or any vague reference to past context that needs resolution before the agent can act.

Analyze 10-K annual filings for public companies using Octagon MCP. Use when extracting key financial metrics, risk factors, business overview, management discussion, and regulatory disclosures from SEC 10-K filings.

Analyze 10-Q quarterly filings for public companies using Octagon MCP. Use when extracting quarterly performance metrics, revenue breakdown, operating margins, segment performance, and interim financial updates from SEC 10-Q filings.

This skill should be used when searching Claude Code session transcripts with semantic understanding. Triggers on queries like "find sessions about X", "when did I work on Y", "search previous conversations". Supports natural language queries with synonym matching.

Focused Signals scout for PostHog setup health. Reads the project's active health issues — the deterministic findings of PostHog's own health checks (no live events, outdated SDKs, missing reverse proxy, absent web vitals, ingestion warnings, failing data-warehouse models, and more) — and decides which are genuinely worth surfacing. Unlike a one-signal-per-issue push, it bundles kind-clusters into a single finding, weights by real blast radius (cross-referencing actual event volume and reach), and prioritizes issues an agent can resolve via the MCP. Emits only above the confidence bar; otherwise writes durable memory and closes out empty. Self-contained peer in the signals-scout-* fleet — picked uniformly at random by the coordinator alongside `signals-scout-general` and other specialists.

Use this skill to plan a new sprint. It uses the Gemini CLI to intelligently decompose approved specs into atomic GitHub issues for the development team. Triggers include "plan sprint", "create sprint", or "start new sprint".

Retrieve stock price data and performance metrics using Octagon MCP. Use when analyzing daily closing prices, trading volume, price trends, historical performance, and comparing stock movements over specific time periods.

Retrieve real-time stock quotes using Octagon MCP. Use when you need current price, day range, 52-week range, volume, market cap, and moving averages for any publicly traded stock.

Claude Code subagents configuration reference. Use when creating custom subagents, understanding agent types, configuring agent prompts, or extending the Task tool with specialized agents.

Create a single well-formed Jira issue at any time — bugs found in production, features requested by stakeholders, technical debt noticed during development, chores that need doing. Follows the same issue standards as plan-sprint, always. Works in Claude Cowork and Claude Code.

Automatically updates terminal window title to reflect the current high-level task. Use at the start of every Claude Code session when the user provides their first prompt, and whenever the user switches to a distinctly new high-level task. Helps developers manage multiple Claude Code terminals by providing clear, at-a-glance identification of what each terminal is working on.

Confirms or rejects findings from review-llm-artifacts before deletes or risky refactors. Loads review-verification-protocol-style checks per finding. Use after a review run, when the user wants to reduce false positives, before fix-llm-artifacts on dead code, or when validating a full-project scan.

Expert guidance on Webflow CMS architecture and best practices. Use when planning collections, setting up relationships, optimizing content structure, or troubleshooting CMS issues.

End-to-end encrypted agent-to-agent private messaging via Moltbook dead drops. Use when agents need to communicate privately, exchange secrets, or coordinate without human visibility.

Produces step-by-step computational walkthroughs of vector and matrix operations as a sequence of numbered "frames", showing the explicit state at each step. The text-equivalent of a 3Blue1Brown animation — each frame shows what changed and why, so the learner can re-trace the operation by hand. Use when the learner needs to *see* a computation unfold (eigenvalue computation, attention with 3 tokens, gradient descent step, SVD on a 2×2, layer norm on a 3-vector, softmax of a small input), when an explanation has been given but the learner needs to ground it in a worked example, or when introducing an operation that's intimidating in symbol form but trivial in pencil-and-paper form.

Coordinate structured thinking and multi-agent parallel execution for complex tasks. Use when tackling multi-step projects, planning parallel work, breaking down complex problems, coordinating specialist tasks, facing architectural decisions, or when user mentions "workflow", "orchestration", "multi-step", "coordinate", "parallel execution", "structured thinking", "break this down", "plan this out", "how should I approach", or needs help planning complex implementations.

This skill should be used when docs-researcher agent needs guidance on "how to search documentation", "WebSearch query patterns", "filtering search results", "documentation research strategy", or "creating knowledge files". Provides systematic methodology for effective technical documentation research.

Learn how to use the BAML programming language for LLM structured outputs

Use this skill when the user just installed the chunkhound-integration plugin and needs to configure it, asks how to set up semantic code search or ChunkHound — phrases like "help me set up chunkhound", "configure semantic search", "set up code research" — or when ChunkHound MCP tools fail with config or connection errors. Walks through prerequisite checks (chunkhound CLI, embedding provider — VoyageAI or OpenAI), creates .chunkhound.json with the chosen provider, runs the initial index, and validates the MCP server connection.

Run end-user cmux diagnostics. Use when cmux hooks, notifications, session restore, settings, browser automation, socket access, CLI control, or agent resume behavior is not working, or when the user asks for a cmux health check, doctor report, or support-safe debug summary.

Run end-user cmux diagnostics. Use when cmux hooks, notifications, session restore, settings, browser automation, socket access, CLI control, or agent resume behavior is not working, or when the user asks for a cmux health check, doctor report, or support-safe debug summary.

Context management specialist for multi-agent workflows and long-running tasks. Use when coordinating multiple agents, preserving context across sessions, or managing complex project state. Focuses on context compression and distribution.

Self-improvement patterns for AI agents to learn from feedback, errors, and successful patterns across sessions

Build professional financial services data packs from various sources including CIMs, offering memorandums, SEC filings, web search, or MCP servers. Extract, normalize, and standardize financial data into investment committee-ready Excel workbooks with consistent structure, proper formatting, and documented assumptions. Use for M&A due diligence, private equity analysis, investment committee materials, and standardizing financial reporting across portfolio companies. Do not use for simple financial calculations or working with already-completed data packs.

Performs comprehensive, multi-layered research on any topic with structured analysis and synthesis of information from multiple sources. Uses file-based research tracking, parallel investigation threads, and context-efficient patterns for deep investigations. ALL MEDICAL CITATIONS FROM PUBMED MCP ONLY.

Analyzes the Agent Alchemy plugin ecosystem to detect dependency issues across all plugin groups

Select and optimize embedding models for semantic search and RAG applications. Use when choosing embedding models, implementing chunking strategies, or optimizing embedding quality for specific domains.

Investigate AI observability clusters — understand usage patterns in AI/LLM traffic, compare cluster behavior, compute cost/latency metrics, and drill into individual traces within clusters.

Generate Salesforce Flows using the MCP tool execute_metadata_action. Use when the user asks to create, build, or generate a flow — including Screen, Autolaunched, Record-Triggered (before/after-save), Scheduled. Also trigger for flow-like requests such as "when a record is created", "trigger daily at", "send an email when", "update the field when", "automate", "workflow", or "flow XML/metadata". This is the only skill for Salesforce Flow generation.

Retrieve historical financial ratings and key metric scores over time using Octagon MCP. Use when analyzing overall ratings, return on assets, return on equity, discounted cash flow scores, debt-to-equity scores, and letter grades (A+, A, B, etc.) for any public company.

LangChain LLM application framework with chains, agents, RAG, and memory for building AI-powered applications

MCP (Model Context Protocol) server build and evaluation guide, including local conventions for tool surfaces, config, and testing

Optimize prompts for clarity and effectiveness. Use when user says "improve this prompt", "optimize my prompt", "make this clearer", or provides vague/unstructured prompts. Intelligently routes to sub-agents for codebase research, clarifying questions, or web search as needed.

Analyze Claude Code and OpenCode sessions for patterns and improvements

Orchestrate parallel deep research across multiple LLM providers using native context:fork subagents and synthesize results

Load top-performing Shinka programs into agent context using `shinka.utils.load_programs_to_df`, and emit a compact Markdown bundle for iteration planning.

Interact with the Stakr protocol — ERC-4626 vaults with multi-reward staking. Use when the user or an agent needs to work with Stakr vaults, add or modify rewards (addRewardToken, addRewards, modifyRewardToken, modifyReward), create or manage an "agent vault" or "own vault", fund incentive programs, configure reward schedules, or integrate Stakr in scripts or tooling. Prefer this skill whenever Stakr, StakrVault, rewards, staking, or vault ownership is mentioned.

Structured messaging protocols for agent team communication including message type selection, plan approval, shutdown procedures, and anti-patterns to avoid. Use this skill when establishing communication norms for a newly spawned team, when deciding which message type to use (broadcast was removed; address recipients individually), when a team-lead needs to review and approve an implementer's plan before work begins, when orchestrating a graceful team shutdown after all tasks are complete, or when debugging why teammates are not coordinating correctly at integration points.

Structured messaging protocols for agent team communication including message type selection, plan approval, shutdown procedures, and anti-patterns to avoid. Use this skill when establishing communication norms for a newly spawned team, when deciding whether to send a direct message or a broadcast, when a team-lead needs to review and approve an implementer's plan before work begins, when orchestrating a graceful team shutdown after all tasks are complete, or when debugging why teammates are not coordinating correctly at integration points.

First-time setup for the Teamcraft JCG plugin. Verifies that the Atlassian MCP server (sooperset/mcp-atlassian) and GitHub CLI are configured and working, guides through setup if not, and recommends companion plugins. Run this before any other Teamcraft skill. Works in Claude Code and Claude Cowork.

Extract a validated learning about Claude Code behavior from the current session and add it to the project's CLAUDE.md memory file. User-only maintenance workflow for updating durable main-agent instructions after a session reveals a rule that should persist.

Use when designing agent work contracts, handoffs, evidence, and role boundaries across runtimes. Triggers:

Use when writing instructions that guide Claude behavior - skills, CLAUDE.md files, agent prompts, system prompts. Covers token efficiency, compliance techniques, and discovery optimization.

Multi-agent orchestration patterns including coordinator agents, LlmAgent vs WorkflowAgent selection, agent-as-tool pattern, and inter-agent communication via session state. PROACTIVELY activate for: (1) multi-agent systems and agent coordination, (2) sub-agent delegation and agent-as-tool implementation, (3) workflow orchestration with LlmAgent and WorkflowAgent. Triggers: "multi-agent", "orchestration", "agent team"

This skill should be used when the user wants to create a custom agent alias theme, edit an existing theme, preview theme mappings, or manage alias configurations. Triggers on phrases like "create a theme", "new alias theme", "custom theme", "edit theme", "preview theme", "show themes", "list themes", "alias", "character theme", "personality theme", "add a theme".

Conduct comprehensive market research using Y Combinator data, SEC filings, social media insights, and web scraping via anysite MCP server. Analyze tech markets, research startup ecosystems, study public companies, identify market opportunities, and understand competitive dynamics. Supports startup discovery, industry analysis, public company research, and social sentiment analysis. Use when users need to analyze market opportunities, research industries, evaluate startups, study public companies, or gather market intelligence for strategic planning and investment decisions.

# Skill: twitter-agent

Guides the agent through migrating an existing Capacitor app project from CocoaPods to Swift Package Manager (SPM) for iOS dependency management. Covers prerequisite checks, inventorying installed Capacitor plugins, backing up customized iOS project files (Info.plist, AppDelegate.swift, Assets.xcassets, Base.lproj, App.entitlements, GoogleService-Info.plist, .xcconfig files, signing configuration), deleting the existing `ios/` folder, re-scaffolding with `npx cap add ios --packagemanager SPM`, restoring preserved files, re-syncing plugins, and verifying the build. Performs all migration steps manually — does not use the interactive `npx cap spm-migration-assistant` command. Do not use for Capacitor plugin projects, app projects already on SPM, app projects without an existing `ios/` folder, or non-Capacitor mobile frameworks.

Set up a persistent, LLM-maintained wiki inside any project. Creates a wiki/ directory with sources, pages, schema, and navigation files, seeds initial pages from project discovery, and injects CLAUDE.md rules that make Claude automatically maintain the wiki during normal work sessions.

Facebook's library for efficient similarity search and clustering of dense vectors. Supports billions of vectors, GPU acceleration, and various index types (Flat, IVF, HNSW). Use for fast k-NN search, large-scale vector retrieval, or when you need pure similarity search without metadata. Best for high-performance applications.

Multi-AI implementation using Codex and Gemini CLIs (Double Diamond Develop phase)

Use when the user asks to run Gemini CLI for code review, plan review, or big context (>200k) processing. Ideal for comprehensive analysis requiring large context windows. Uses Gemini 3 Pro by default for state-of-the-art reasoning and coding.

Use Gemini CLI in non-interactive mode for tasks requiring massive context windows (1M tokens). Best delegated to subagents for iterative analysis and summarization. Invoke when analyzing large codebases, requesting deep analysis, getting second opinions on complex problems, or when Claude's context limits are insufficient. Triggers include phrases like "use gemini", "analyze with gemini", "get second opinion", "deep analysis of codebase", or when processing files exceeding Claude's context capacity. IMPORTANT - Always delegate to a subagent using the Task tool for better iteration and result summarization.

Create and maintain GitHub Agentic Workflows (gh-aw) for Prowler. Trigger: When creating agentic workflows, modifying gh-aw frontmatter, configuring safe-outputs, setting up MCP servers in workflows, importing Copilot Custom Agents, or debugging gh-aw compilation.

Review AI-generated code changes before committing using GitHuman. Use when reviewing code changes, creating code reviews, checking what the AI agent wrote, preparing to commit, or when user mentions "review", "GitHuman", or "before commit".

Audit and trim Claude Code instruction setups that have grown bloated over time. Use when outputs feel worse, rules have accreted across CLAUDE.md/skills/context files, or you want a recurring setup hygiene check.

Extract insights from multi-agent interactions, identify patterns, and build collective intelligence through cross-agent learning and knowledge management. Use when synthesizing findings, building knowledge bases, or improving system-wide practices.

This skill should be used when the user asks to "share memory between agents", "KV cache compaction for multi-agent", "orchestrator worker context", "latent briefing", "reduce worker tokens", "cross-agent memory without summarization", or discusses Attention Matching compaction, recursive language models with workers, or token explosion in hierarchical agents.

This skill should be used when the user asks to "share memory between agents", "KV cache compaction for multi-agent", "orchestrator worker context", "latent briefing", "reduce worker tokens", "cross-agent memory without summarization", or discusses Attention Matching compaction, recursive language models with workers, or token explosion in hierarchical agents.

Implement comprehensive evaluation strategies for LLM applications using automated metrics, human feedback, and benchmarking. Use when testing LLM performance, measuring AI application quality, or establishing evaluation frameworks.

Prepare structured briefings for meetings with legal relevance and track resulting action items. Use when preparing for contract negotiations, board meetings, compliance reviews, or any meeting where legal context, background research, or action tracking is needed.

Use this skill when integrating, reading, or updating a project's conventions.md file — the document that captures house style for issue titles, labels, branch naming, PR descriptions, handoff phrasing, and review etiquette. Trigger phrases "set conventions", "load conventions", "where are the conventions", "what are our conventions", "add a convention", "edit conventions.md", "/pm-conventions", "the conventions say", "our team uses <X> format", "project conventions", "project title format". Also use proactively when another pm-* skill (pm-issues, pm-review, pm-improve, pm-branches, pm-handoff, pm-agent, pm-projects, pm-project-review) is about to write or evaluate text and a conventions file exists at `<repo>/.claude/pm/conventions.md` — the conventions must be loaded and applied before the skill writes anything user-facing.

Generates and maintains a hierarchical .context/ directory with architecture overviews, coding conventions, domain glossaries, and dependency maps that give any LLM working in the project the right background. Use when setting up a new project for AI-assisted development, when the user asks to create or update context docs, when onboarding a new LLM tool to an existing codebase, or when the user says the AI doesn't understand their project.

A/B testing and performance metrics for prompts

Extend Pydantic AI agents with batteries-included capabilities from pydantic-ai-harness — currently Code Mode, which collapses many tool calls into one sandboxed Python execution. Use when the user mentions pydantic-ai-harness, CodeMode, Monty, code mode, or tool sandboxing, when they want an agent to run agent-written Python, or when a Pydantic AI agent would benefit from orchestrating multiple tool calls in a single sandboxed script.

Vertex AI RAG Engine integration patterns for grounding agent responses in private data sources including corpus management, retrieval tool creation, and citation extraction. PROACTIVELY activate for: (1) RAG pipeline integration and Vertex AI RAG Engine setup, (2) corpus creation and document ingestion, (3) retrieval tool configuration and grounding metadata parsing. Triggers: "add rag", "rag pipeline", "vertex ai search"

Create and maintain agent rules in AGENTS.md and .agents/rules/. Use for project rules, conventions, constraints, rule indexes, or requests to add or optimize agent rules.

Sets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`, `triage`, `diagnose`, `tdd`, `improve-codebase-architecture`, or `zoom-out` — or if those skills appear to be missing context about the issue tracker, triage labels, or domain docs.

Sets up an `## Agent skills` block in AGENTS.md/CLAUDE.md and `docs/agents/` so the engineering skills know this repo's issue tracker (GitHub or local markdown), triage label vocabulary, and domain doc layout. Run before first use of `to-issues`, `to-prd`, `triage`, `diagnose`, `tdd`, `improve-codebase-architecture`, or `zoom-out` — or if those skills appear to be missing context about the issue tracker, triage labels, or domain docs.

Create new Claude Code skills following project patterns and best practices. Use when building new skills, extracting reusable capabilities, or converting commands to skills.

Iteratively reviews and fixes Claude Code skill quality issues until they meet standards. Runs automated fix-review cycles using the skill-reviewer agent. Use to fix skill quality issues, improve skill descriptions, run automated skill review loops, or iteratively refine a skill. Triggers on 'fix my skill', 'improve skill quality', 'skill improvement loop'. NOT for one-time reviews—use /skill-reviewer directly.

Generate complete academic survey papers using multi-LLM parallel outline generation, RAG-based subsection writing, citation validation, and local coherence enhancement. Based on AutoSurvey pipeline. Use for writing comprehensive literature surveys.

Tidewave MCP dev tools for Phoenix applications. Use when setting up Tidewave, connecting AI coding assistants to a running Phoenix app, configuring MCP server access, debugging with runtime introspection tools, or troubleshooting Tidewave integration.

Maintain a structured task list for the current session. Use proactively when a request requires 3+ distinct steps, the user provides multiple items, or complex work benefits from explicit progress tracking. Storage persists via Muninn config across container death. Adapted from Claude Code's TodoWrite tool.

WCAG 2.2 AA learning and reference agent. Use when you need to understand WCAG success criteria, learn what changed between versions, understand conformance levels, clarify when criteria apply, or get plain-language explanations of accessibility standards. Does not write or review code - teaches the standard itself.

Deliberate fresh-eyes self-review and repair after making changes. Use when an agent has just written or modified code, config, tests, or docs and should pause to look for obvious bugs, regressions, missing tests, confusing behavior, or risky assumptions, fix the clear local issues it finds, and only then finalize, hand off, or commit. Supports Claude Code, Codex, and Gemini with provider notes in references/.

Bootstrap a new AI-agent session by reading the repository operating docs and rebuilding project understanding from source. Use when Codex needs to start a new conversation, get up to speed on an unfamiliar repo, read AGENTS.md and README.md completely before acting, investigate the codebase to understand the project's purpose and architecture, or turn a rough onboarding prompt into an execution-ready repo-orientation prompt.

This skill should be used when the user asks to "keep cache warm", "warm the cache", "save tokens", "reduce token usage", "cache-money", "start cache ping", "optimize peak hours", "prevent cache expiry", or mentions Anthropic prompt caching, cache TTL, peak hour throttling, token cost optimization, or keeping the context cache alive in Claude Code sessions.

Guides the agent through Angular-specific patterns for Capacitor app development. Covers project structure, adding Capacitor to Angular projects, using Capacitor plugins in Angular services and components, NgZone integration for plugin event listeners, lifecycle hook patterns, dependency injection, routing with deep links, and environment-based platform detection. Do not use for creating a new Capacitor app from scratch, upgrading Capacitor versions, installing specific plugins, Ionic Framework setup, or non-Angular frameworks.

This skill is the mandatory reference for all Claude Code hook creation. Use cc-hooks-ts for every hook. This skill should be used when the user asks to "create a hook", "add a hook", "write a hook", "implement a hook", "rewrite hooks in TypeScript", "use cc-hooks-ts", or needs to build any Claude Code hook — cc-hooks-ts is always required regardless of whether explicitly mentioned.

Prime the agent with a behavioral persona for the conversation

Use the ctx7 CLI to fetch library documentation, manage AI coding skills, and configure Context7 MCP. Activate when the user mentions "ctx7" or "context7", needs current docs for any library, wants to install/search/generate skills, or needs to set up Context7 for their AI coding agent.

Customer Relationship Management (CRM) vertical skill for the Customware SPA. Defines the section layout, entity views, and mapping rules for transforming a DOMAIN.md into a CRM tool. Use this skill when the Builder Agent classifies a customer's domain as contact management, sales pipeline, deal tracking, or customer relationship workflows. Trigger signals: contacts, leads, deals, pipeline, follow-ups, opportunities, accounts, customer data everywhere, can't track who talked to whom, sales process, referrals.

Conducts deep, multi-angle research using glim MCP tools and parallel subagents. Use for deep research, competitive landscape analysis, strategic intelligence, or /deep-research-glim [topic]. Triggers - deep research, deep dive on, competitive landscape, strategic intelligence, multi-source synthesis.