AI & ML

heypinchy Skill

Create and update records in a connected Odoo instance with odoo_create / odoo_write, following the standard write discipline. Use whenever an agent has write access and needs to add or change data. Covers draft-first and confirm-before-write, duplicate checks, single-call creation with inline child lines, never guessing relation IDs, and verifying the record after create.

heypinchy Skill

Search the public web for current information and fetch full pages. Use when the user asks about current events, recent company news, market data, regulatory changes, or any fact that may have changed since the model's training cutoff.

heypinchy Skill

Attach files (receipts, delivery notes, contracts, supporting documents) to Odoo records with odoo_attach_file using the _pinchy_ref token. Use whenever the user sends a document that belongs on an Odoo record. Explains the ref-flow contract so you pass a real _pinchy_ref instead of fabricating a "<model>,<id>" string.

heypinchy Skill

Schedule, complete, and reschedule Odoo activities (the "needs attention" follow-ups on any record) with the odoo_schedule_activity / odoo_complete_activity / odoo_reschedule_activity tools. Use whenever an agent adds, closes, or pushes a follow-up on a lead, order, ticket, request, or any record. Explains why you must never odoo_create / odoo_write on mail.activity directly.

heypinchy Skill

Convert gross (tax-inclusive) totals from receipts and invoices into the tax-exclusive net price_unit Odoo stores on account.move.line. Use whenever an agent books a bill or invoice from a document that shows gross totals. Covers the per-line net formula, multi-line splits, the rounding tolerance, and overriding a line's default tax.

heypinchy Skill

Disambiguate Odoo accounting records across multiple companies. Use whenever an agent reads or writes ledger records (account.move, account.account, account.journal, account.tax, account.payment) on an instance that may host several companies. Covers reading the [Company X] label suffix, filtering by company_id, resolving cross-company multi-match errors, and setting company_id on creates.

heypinchy Skill

Find and read the documents an agent has been granted access to, using pinchy_ls and pinchy_read. Use whenever a question is about the content of a file in the agent's document folders. Covers discovering files before reading them, which formats can actually be extracted, how to point a reader back at a passage, and what to do when a document is missing or unreadable.

htkoca Skill

CLAUDE.md, AGENTS.md, .claude/, agent instructions file, repo hygiene for Claude Code, gitignoring local agent state, settings.local.json. Use when setting up a repo for Claude Code, writing or restructuring a CLAUDE.md, auditing a repo that has both CLAUDE.md and AGENTS.md, or deciding what agent state gets committed.

managedcode Bundle

Create or update an ADR under `docs/ADR/` for architectural decisions, dependency changes, data-model changes, or cross-cutting policy shifts. Use when the user asks to write, update, or document an ADR, record a design decision, capture architecture trade-offs, or justify a repo-wide technical policy.

managedcode Bundle

Apply ML/AI project delivery guidance for data exploration, feasibility, experimentation, testing, responsible AI, and operating ML systems. Use when the repo includes model training, inference, data science workflows, or ML-specific delivery planning.

managedcode Bundle

Set up or refine solution-level governance for MCAF repositories: root and project-local `AGENTS.md`, rule precedence, solution topology, skill routing, and maintainability-limit policy placement. Use when bootstrapping a repo, restructuring a multi-project solution, or tightening agent rules.

marketer-gq Skill

Draft, revise, and review the participant-facing emails on a Great Question study — invitations, reminders, and post-session messages — with your team's voice, mandatory disclosures, and incentive-mention rules applied consistently. Use when writing or improving the emails participants receive; the skill never sends without confirmation. Requires the Great Question MCP integration to be connected.

marketer-gq Skill

Find the right candidates for a Great Question study, filter for fit, shortlist them, and send screener invitations, applying your org's recruitment rules. Use when recruiting participants onto a study by name, by segment, or by a description of who you're looking for; the skill confirms before anything is sent. Requires the Great Question MCP integration to be connected.

marketer-gq Skill

Audit your Great Question candidate panel for over-recruitment, thin segments, and cooldown violations, then propose remediation. Read-only — it reports and recommends but never modifies candidates. Use before a recruitment-heavy study or as a regular panel health check. Requires the Great Question MCP integration to be connected.

marketer-gq Skill

Set up a complete moderated interview study in Great Question from a research brief — study, screener, moderators, scheduling, livestream, and incentive — in one orchestrated workflow, stopping just short of activation. Use when a researcher wants to launch a new interview study and has a brief or enough context to describe one. Requires the Great Question MCP integration to be connected.

marketer-gq Skill

Set up a complete unmoderated test in Great Question from a research brief — study, blocks, screener, and incentive — in one orchestrated workflow, stopping just short of activation. Use when launching a prototype test, website test, card sort, tree test, or survey-style task collection. Requires the Great Question MCP integration to be connected.

martintmk Bundle

Review a Rust pull request, branch, commit or working-tree diff as an autonomous AI reviewing agent applying @martintmk's library-maintainer priorities. Establishes facts once, dispatches every review sub-skill to a fresh agent context, merges findings and delivers one AI-attributed review. Use for "review this PR", "review my changes" or "review like me". For a focused area, invoke its review-* skill directly. Not for formatting-only passes, output-only API audits or specialist security reviews.

c-kick Skill

Browse and visually inspect webpages using an available browser automation tool, preferably headless Chromium via Puppeteer MCP. Use when you need to see a page, interact with a web UI, extract information that a plain fetch cannot get (JS-rendered content, SPAs, authenticated pages), or when the user wants to show you something in a browser.

jawwad-ali Skill

This skill should be used internally to decide when to use curated skills versus Context7 MCP for documentation lookups. Use this when handling technical questions about frameworks, libraries, or APIs to determine the optimal knowledge retrieval strategy.

jawwad-ali Bundle

This skill should be used when the user asks to "create an AI agent", "build a multi-agent system", "implement agent handoffs", "add guardrails to agents", "create function tools for agents", "implement agent tracing", "build agentic workflows", "create a triage agent", "implement input/output guardrails", "add tools to agents", or mentions OpenAI Agents SDK, agents, handoffs, guardrails, Runner, or multi-agent orchestration.

lushows Bundle

optimizer_tokens_lushows — Tu experto en economía de tokens LLM

bitsocialnet Skill

Investigate a reported bug using source, tests, or runtime evidence to establish its cause.

bitsocialnet Skill

Investigate a reported bug using source, tests, or runtime evidence to establish its cause.

lev-os Skill

Multi-agent orchestrator for Claude Code. Use when user mentions gastown, gas town, gt commands, bd commands, convoys, polecats, crew, rigs, slinging work, multi-agent coordination, beads, hooks, molecules, workflows, the witness, the mayor, the refinery, the deacon, dogs, escalation, or wants to run multiple AI agents on projects simultaneously. Handles installation, workspace setup, work tracking, agent lifecycle, crash recovery, and all gt/bd CLI operations.

lev-os Skill

Agent-human interaction protocol for AI agents. Use when agents need human approvals, selections, research direction, or async feedback. Triggers on "human in the loop", "approval", "agent interaction", "get human input", or "wait for response".

lev-os Skill

Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.

lev-os Skill

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.

iliaal Bundle

Transforms vague or rough prompts into precise, structured AI instructions. Use when asked to "refine prompt", "improve prompt", "make this prompt better", "promptify", "optimize prompt", "rewrite prompt", "enhance prompt", or "sharpen instructions".

drmoisan Skill

Execute a previously planned epic by replaying the committed epic-kickoff artifact through the epic-orchestrator agent. Use after /epic-plan has prepared the epic (issues, folders, research, specs, atomic plans, preflight clearance) and the user is ready to execute end-to-end.

drmoisan Skill

Scope and prepare a multi-feature epic end-to-end before execution for the epic-planner agent - epic-worthiness gate, decomposition, dependency-wave design, complexity assessment, preparation-mode child orchestrator delegations, integration-branch fan-in, and the epic-orchestrator kickoff prompt artifact.

drmoisan Skill

Route a multi-feature epic through the deterministic wave-scheduling, integration-branch, and fan-in workflow for the epic-orchestrator agent.

neuralblitz Skill

Expert guidance on Ray distributed computing framework for scaling AI and Python applications. Use for: distributed training, hyperparameter tuning with Ray Tune, reinforcement learning with Ray RLlib, model serving with Ray Serve, batch inference, scaling Python applications from single machines to clusters, and Ray Core tasks and actors.

neuralblitz Skill

Expert assistant for working with Anthropic's Claude API and Claude-powered applications. Use this skill whenever you need help with: API integration, message formatting, tool use implementation, context window management, Constitutional AI principles, prompt engineering, Claude models (Claude 3.5, Claude 3), vision capabilities, computer use, or building production systems with Claude. Also use for understanding Anthropic's safety approaches, token optimization, or any Claude-related development task.

ysyecust Skill

複数のストレージレイヤー（ローカルファイル、MCP メモリ、ベクターストア、Git リポジトリ）にわたるナレッジベースの管理、取り込み、同期、検索。ユーザーが知識システム全体で保存・整理・同期・重複排除・検索を行いたい場合に使用します。

gavishap Skill

Answer questions from the compiled Wiki. Use whenever the user asks a knowledge question that this vault might cover.

gavishap Skill

Compile a Raw source into concise, linked Wiki notes. Use when a new source lands in Raw/Sources/ or when a source is marked Processed:false.