The library

localization-manager · See body for use case.

political-firehose · See body for use case.

type_checker · Researchers validating GNN model schemas and auto-generating comparison cards with FLOPS.

portfolio-audit · Fund managers auditing multi-position books for concentration and stress-test exposure in minutes.

agentdesk · Teams running collaborative AI agent sessions with persistent context and quality gates.

llm-context · Teams or operators requiring llm-context for skill-level automation.

price-alert · Traders setting hands-off alerts across a watchlist that fire via Telegram 24/7.

implementing-code · Teams or operators requiring implementing-code for skill-level automation.

review-investment-screenshot · Quick due diligence on investment screenshots—technicals, catalysts, profit-taking rules applied.

dev-workflow · Teams shipping code with strict quality gates, test-first discipline, and audit trails.

sector-rotation-analysis · Allocators identifying overheated vs undervalued sectors and executing tax-aware rotation pairs.

strategic-partner-firehose · Investors finding Tier-1 strategic PIPE deals from SEC filings before social media coverage.

claude-switch · Users managing multiple Claude Code accounts without re-authentication overhead.

summarize · Teams or operators requiring summarize for skill-level automation.

tax-optimize · Teams or operators requiring tax-optimize for skill-level automation.

claude-code · Teams or operators requiring claude-code for skill-level automation.

building-project · Teams or operators requiring building-project for skill-level automation.

agent-workflow · Teams running collaborative AI agent sessions with persistent context and quality gates.

githits · Teams or operators requiring githits for skill-level automation.

doccraft-adr · Teams or operators requiring doccraft-adr for skill-level automation.

email-send · Teams or operators requiring email-send for skill-level automation.

doccraft-close · Teams or operators requiring doccraft-close for skill-level automation.

electronic · When Prepare CP2K electronic-analysis task inputs from prior converged context. Use when the user requests post-ground-state electronic analyses and needs prerequisite-aware setup

doccraft-config · After doccraft init: run Analyse mode to tailor the freshly scaffolded

camera-shake · When Shake the camera when the player takes damage or on impacts for visual feedback

doccraft-queue-audit · - After adding a story, changing depends_on / id, or

doccraft-session-wrap · - The user invokes this skill or asks to wrap this session /

firecrawl-scraper · - When you need deep content extraction from web pages

doccraft-story · - Creating a new story in docs/stories/.

doccraft-update · - Periodically to keep doccraft skills and openspec instructions current.

openspec-apply-change · When Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks

sverklo · When sverklo capability is needed

openspec-archive-change · When Archive a completed change in the experimental workflow. Use when the user wants to finalize and archive a change after implementation is complete

openspec-explore · When Enter explore mode - a thinking partner for exploring ideas, investigating problems, and clarifying requirements. Use when the user wants to think through something before or during a change

haus-workflow · When haus workflow capability is needed

ai-fob · When ai fob capability is needed

sql-escaper · When sql escaper capability is needed

pi-dynamic-workflows · When pi dynamic workflows capability is needed

nexo-brain · When nexo brain capability is needed

doccraft-adr · - A chat or design exploration reached a conclusion worth keeping (yes, no, defer, or "use X instead of Y").

doccraft-close · - The user wants to close / ship / finish / resolve a specific story and

doccraft-config · After doccraft init: run Analyse mode to tailor the freshly scaffolded

doccraft-queue-audit · When you need to detect blocked work and identify what can run in parallel after adding or editing dependencies.

doccraft-session-wrap · When a design or research thread concludes and you want to freeze decisions into durable docs without manual drafting.

whyline · When you want to preserve the why behind code changes so future context windows recall intent without re-reading chat.

doccraft-story · When you need a structured, machine-readable backlog item linked to dependencies and roadmap rows.

harnessed · When you want structured three-layer-stack workflow without learning each upstream terminology or manually routing stages.

adversarial-workflow · When you want battle-tested output that survives independent scrutiny rather than single-agent rubber-stamping.

tmr-shopify-store-audit · When you need a premium ecommerce CRO audit with visual clarity and ranked action items instead of raw CSV export.

anthropic · When integrating Claude calls into a Genkit pipeline or you need prompt caching and citations for reliability.