Aggregate and deduplicate recent news from multiple sources into concise topic summaries.

```md

OAuth 2.0 and OpenID Connect expert for authorization flows, PKCE, and token management

OAuth and OIDC misconfiguration testing playbook. Use when reviewing redirect URI handling, state and nonce validation, PKCE, token audience, callback binding, and identity-provider trust flaws.

Obsidian vault: search/read/write notes, backlinks, Bases, Canvas.

Use for first-time SiftCoder setup or onboarding a new user/project. Walks through Ollama probe, memory daemon start, scope setup, optional backfill. Opinionated, single happy path.

Use for performance work. Measure-first, profile-driven. Refuses to optimise without a baseline measurement. Targets the actual hotspot, not where intuition says it is.

Use when the user wants predictive intent inference — "what am I about to do", "what would I likely want next", "predict the next step". Mines memory patterns to forecast the next action.

Use when the user wants to pair-program with approval at each step. AI proposes one change at a time; user accepts/edits/rejects each before next move. Slow, deliberate, high-trust mode.

Create and manage git worktrees for parallel coding sessions with zero dead time. Use when blocked on tests, builds, wanting to work on multiple branches, context switching, or exploring multiple approaches simultaneously.

Use when the user wants to extract a reusable pattern from recent work and store it for future use. "Save this as a pattern", "remember this approach", "learn from this commit".

Searches memories and displays compact one-liner results, or looks up a specific memory by ID. Use for quick memory lookups, checking if a decision was recorded, resolving [mem0:id] citations, or browsing memories without full category detail.

Analyze permission denial patterns and generate optimized alwaysAllow and alwaysDeny rules. Use when permission prompts are slowing you down or after sessions with many denials.

Use when motion needs to feel physically grounded—objects falling, characters jumping, things colliding, or any element that should obey believable weight and momentum.

# Pi-hole Skill

Manage plans and context documents with automatic indexing

Use Trekker to plan and track multi-step tasks BEFORE starting implementation.

Use when creating animations that entertain, engage with humor, or create lighthearted interactive experiences.

Optimize pyproject.toml and resolve complex dependency trees using modern tools like Poetry or uv. Use to modernize Python project management.

Postman CLI reference and git sync file structure knowledge - provides context for CLI-based commands (send-request, generate-spec, run-collection, context)

Reviews pull requests or changesets to ensure only minimal, necessary changes are included, with no extraneous modifications or artifacts.

Merge multiple PRs into a temporal integration branch before merging to base, with ordered conflict resolution. Use when you want to validate a set of PRs together on a staging branch before advancing the base branch.

Use when the user wants diff-before-apply discipline — show what would change, get approval, then apply. Risk-area enforcement. "Preview the change", "show before applying".

Prioritize a backlog of feature ideas based on impact, effort, risk, and strategic alignment with top 5 recommendations. Use when prioritizing a feature backlog, making scope decisions, or ranking product ideas.

Execute an implementation plan with surgical precision. Use after a planning phase (plan-now or similar) has produced a step-by-step strategy and identified critical files. Focuses on precise code changes with verification at each step.

Use when prioritizing animation features, building motion roadmaps, or when a PM needs to understand the business value of animation principles.

# Production Engineering Skill

Use when creating animations for business contexts that require seriousness, competence, and trustworthy presentation.

This skill provides project-specific context including project nature, technology stack, and implementation principles. Automatically loaded to understand project constraints and conventions. Should be customized for each project via /setup-context.

# Skill: Project Manager (Vivi OS)

Project management expert for Agile, estimation, risk management, and stakeholder communication

Track and report project-runner project status. Use when the user asks about project progress, status, milestones, or tasks for an active project-runner project. Reads the project's status.md and milestone files to provide a comprehensive progress report.

Detect project type and route to the appropriate publication workflow. Routes R packages to the CRAN/JOSS pipeline, Python packages to PyPI publishing, or asks for clarification if ambiguous.

Generate and validate environment-based configuration for Python apps using Pydantic or Dynaconf. Use to ensure secure and valid runtime settings.

Analyze and resolve Python package dependency conflicts. Use when pip install fails due to version mismatches or circular dependencies.

Enforce Pythonic standards using Black, Isort, and Flake8. Use to ensure consistency across large Python codebases and team environments.

Identify CPU and memory bottlenecks in Python code using cProfile or memory_profiler. Use to optimize mission-critical Python services.

Automatically add or improve type annotations in legacy Python code. Use to improve code readability, IDE support, and catch type errors early.

Setup and validate Python virtual environments (venv, virtualenv, conda). Use to ensure isolated dependencies and correct Python versions for projects.

AHRS quaternion decomposition and roll/pitch compensation utilities. Use when working with pywayne.ahrs.tools module for quaternion to Euler angle decomposition, or compensating orientation to zero pitch and roll using roll-pitch compensation method.

Guides Qdrant query latency optimization. Use when someone asks 'search is slow', 'how to reduce latency', 'p99 is too high', 'tail latency', 'single query too slow', 'how to make search faster', or 'latency spikes'.

Search the FPF knowledge base and display hypothesis details with assurance information

R plotting conventions for Seurat/scRNA-seq: UMAP square plots (aspect.ratio=1, raster=FALSE), polychrome cluster colors, DimPlot/FeaturePlot settings, saving 300 DPI PNG, volcano plots with EnhancedVolcano, statistical comparisons with rstatix/ggpubr.

Explain the Ralph Loop plugin, how it works, and available skills. Use when the user asks for help with ralph loop, wants to understand the technique, or needs usage examples.

Guide a safe git rebase of the current branch onto a target branch, including conflict triage and resolution steps. Use when asked to rebase, update a branch, or resolve rebase conflicts.

Use when completing a plan, finishing a development branch, wrapping up a session, or at any natural transition between work phases — reviews skill-bus telemetry to identify subscription gaps and suggest improvements

Bayesian regression models including linear, logistic, Poisson, negative binomial, and robust regression with Stan and JAGS implementations.

Stores a memory verbatim from user input with appropriate type classification and metadata. Use when the user says remember this, save this, store this, note that, or explicitly asks to record a decision, preference, convention, or learning.

Remote Macs: MacBook, Mac Studio, clawmac, Tailscale, SSH, OpenClaw.

Repository information using GitHub CLI. Trigger when user wants to view repository details ("show repo info", "what's this repo about"), list repositories ("list my repos", "show repos for user"), or check repository metadata ("stars", "languages", "topics").

Report a bug in the compound-engineering plugin

Plan and optimize resource allocation. Trigger with "resource planning", "capacity", "utilization", "staffing plan", "who should work on what", "we're stretched thin", or when the user needs help allocating people, budget, or time across projects and teams.

MUST use when managing resources.

Rewrite branch into clean, narrative-quality commits. Creates backup, reimplements on fresh branch, verifies byte-identical, then replaces original branch history.

Identify, assess, and mitigate operational risks. Trigger with "what are the risks", "risk assessment", "risk register", "what could go wrong", or when the user is evaluating risks associated with a project, vendor, process, or decision.

Use when implementing Disney's 12 animation principles with Rive interactive animations

Run AI models on Replicate via predictions, webhooks, and streaming.

Generate database schemas from descriptions

Use to define/show/modify the .siftcoder/scope.json file that the boundary-enforcer hook reads. Bounds where AI can write. "Limit to these files", "lock down scope", "show current scope".

Use when triggering animations on scroll - reveal effects, parallax, sticky headers, progress indicators, or any scroll-linked motion.

Use for codebase-wide search across code + memory + docs simultaneously. Different from mem_search (memory-only) and Grep (code-only). One query, three sources, deduplicated ranking.

BM25 full-text search across all D/P/T/I/H/W/K documents in .crabshell/. Returns ranked results by relevance. Use when looking for specific discussions, plans, tickets, or investigations by topic keyword. Invoke with /search-docs <query>.

Sequencing read requirements and bcl2fastq base masks for 10x Genomics libraries: chemistry-specific BasesMaskCR values, ATAC I2 barcode requirement (capital I not lowercase y), Visium FFPE vs frozen R2 length difference.

Use when identifying bottlenecks, waste, and failure modes in a service — designs PDSA improvement experiments with lane-anchored measures

Use post-session to evaluate what worked, extract reusable patterns, capture lessons. Feeds knowledge base. "What did we learn", "evaluate this session".

Generate a structured handoff document capturing current progress, open tasks, key decisions, and context needed to resume work. Use when ending a session, saying "continue later", "save progress", "session summary", or "pick up where I left off".

# Session Manager Skill

This skill should be used when the user asks to "create a bash script", "write a shell script", or mentions shell scripting conventions.

Complete guide to implementing simple forms (action-only forms without input fields) in finstreet/boilerplate applications. Use when building or modifying any simple form.

Instructions for simplifying code.

Clear description of what this skill does and when to use it. Include trigger keywords and contexts inline, e.g. "Use when user wants to X, Y, or Z."

Create and edit presentation slides using Slidev framework when user requests slides, presentations, or deck modifications

Use when building deliberate motion between 800-1200ms - dramatic reveals, loading sequences, storytelling moments where users should pause and absorb

Use when building small transitions between 200-300ms - modal appearances, card expansions, navigation transitions that users consciously perceive

Generate feature specifications by analyzing existing source code.

Generate or regenerate specs/OVERVIEW.md with a tabular summary, status, and per-spec file listings

Plan a sprint with capacity estimation, story selection, dependency mapping, and risk identification. Use when preparing for sprint planning, estimating team capacity, selecting stories, or balancing sprint scope against velocity.

SQLite expert for WAL mode, query optimization, embedded patterns, and advanced features

Use when implementing deformation effects, bounce animations, impact responses, or any motion requiring organic elasticity and weight expression.

SSH remote access patterns and utilities. Connect to servers, manage keys, tunnels, and transfers.

Use when mapping all actors in a service ecosystem — who serves whom, dependencies, interest alignment and conflict, power dynamics

Use when elements transform in place - toggle switches, expanding accordions, checkbox animations, button states, or any transformation without entering/exiting.

# Statistical Analysis

Runs a healthcheck of the Crabshell plugin state. Reports memory, regressing, verification, counter values, and plugin version with ✓/!/✗ indicators. Invoke with /status.

Use when executing implementation plans with independent tasks in the current session

Identify and validate hidden assumptions before they become bugs. Use before implementing features or when reviewing generated code.

Analyze your reading history and tell you something surprising you don't know about yourself

Use when diagnosing bugs, failing tests, build failures, unexpected behavior, flaky tests, regressions, or production incidents. Requires reproduction, root cause evidence, one-hypothesis fixes, and regression verification before editing.

Systematic feature planning workflow - use when starting complex features requiring structured approach

Systematic feature planning workflow - use when starting complex features requiring structured approach

Use for team-knowledge sharing patterns — share captured patterns/decisions across teammates' SiftCoder installs, set up shared workspaces, federation hygiene.

Diagnose Telegram configuration and connectivity issues. Use when user reports problems with Telegram sync, connection errors, or wants to troubleshoot.

Initialize Telegram configuration. Use when user wants to set up, configure, or connect their Telegram account for the first time.

Bayesian time series models including AR, MA, ARMA, state-space models, and dynamic linear models in Stan and JAGS.

# tmux Workflow Guide

Turn the current conversation context into a PRD and publish it to the project issue tracker. Use when user wants to create a PRD from the current context.