DevOps & Infra

knuckles-team Bundle

Draft caption copy and render it composed onto a visual asset, then run a final readiness check. Use when a user has supplied a topic/reference and voice constraints and wants a review-ready meme image; this workflow never posts or schedules it.

knuckles-team Bundle

Outline and draft a video script, produce the edited video asset, and run a final readiness check. Use when a user has supplied a topic, audience, and voice constraints and wants a review-ready video; this workflow never publishes or uploads it.

knuckles-team Bundle

Gather publicly disclosed ESG (environmental, social, governance) signals for a supplied company universe and structure them into a normalized, validated dataset. Use when a researcher needs raw ESG disclosure data assembled for downstream screening; this workflow does not compute a proprietary ESG score or rating.

knuckles-team Bundle

Run every health-relevant atomic skill this catalog owns — hardware profile, current resource utilization, compose configuration drift, and TLS expiry — in parallel across the fleet, then consolidate the findings into one formatted report. Use when an operator wants a single point-in-time snapshot combining every owned infrastructure health signal; this workflow is read-only.

knuckles-team Bundle

Locate publicly reported macroeconomic indicator releases (CPI, GDP, rates, and similar) and ingest them into the Timeseries Memory backend. Use when a researcher needs a structured macro-indicator time series as input to another pipeline; this workflow does not interpret or forecast regime shifts.

knuckles-team Skill

Provision the agent-utilities database environment end-to-end from credentials: push the ontology to Stardog (prod) or a local SPARQL endpoint (dev), stand up a Postgres carrying Apache AGE + pgvector + ParadeDB pg_search, wire the durable backend, and backfill graph relationships into AGE. Resolves secrets from OpenBao/Vault (with .env fallback). Prompts for environment (prod/dev) and Postgres mode (combined image vs existing/managed). Use when the user wants to "set up Stardog", "host my ontology over SPARQL", "install pg-age/pgvector/ paradedb", "backfill into AGE", "connect my postgres", or "set up my database environment". Do NOT use for OpenBao unseal/seed alone — use secret-vault-manager; or generic container deploys — use docker-compose-operator.

knuckles-team Bundle

Universal Knowledge Assimilation Engine — the "Master Control Program" for autonomous self-evolution. Discovers content from X, ScholarX, and GitHub, classifies it via the UniversalKnowledgeClassifier, ingests high-value content into the Knowledge Graph, and triggers comparative-analysis + SDD generation when evolution potential is detected. Extends the pull-based agent-utilities-evolution skill with push-based evolution from incoming content. Triggers on "assimilate knowledge", "knowledge assimilation", "ingest and evolve", "push-based evolution", "discover and ingest", "x research pipeline", "self-evolve from X", "run assimilation".

knuckles-team Bundle

Crawls web news, evaluates social posts sentiment spikes, fuses sentiment signals, and backtests correlation results.

knuckles-team Bundle

Discover, fetch, and normalize non-traditional (web/news-sourced) market-data signals by composing the catalog's research and finance atomic skills. Use when a quant researcher needs alternative data ingested and quality-checked before using it in a factor or backtest pipeline; this workflow does not place trades or score signal predictiveness.

knuckles-team Bundle

Scaffolds standard-compliant Python strategy files for Freqtrade, runs parallel hyperparameter optimizations, dry-runs them, and elevates to live webhook monitoring.

knuckles-team Bundle

Collects pricing indicators, extracts structural features, fits Hidden Markov Model classifiers, and adapts strategy params to structural shifts.

knuckles-team Bundle

Locate publicly reported earnings-announcement dates for a supplied ticker universe and ingest them into the Timeseries Memory backend. Use when a researcher needs a structured earnings calendar as input to another pipeline; this workflow does not forecast earnings surprises or trade around them.

knuckles-team Bundle

Self-deploying bootstrap workflow that unfolds the complete agent-utilities ecosystem on a fresh machine. Resolves the canonical deployment profile (tiny/single-node-prod/enterprise), installs system dependencies via systems-manager-mcp, deploys core infrastructure (Docker, Portainer, Technitium DNS), service containers (wger, mealie, Jellyfin, etc.), and all MCP server containers as streamable-http Docker deployments. Transitions from container-manager-mcp to Portainer once deployed. Registers the full deployment topology in the Knowledge Graph.

knuckles-team Bundle

Research, outline, draft, verify, edit, and preflight a batch of sourced content pieces by composing the catalog's atomic research and content skills, unattended. Use when a user has supplied a topic list, audience, objective, and publication requirements and wants review-ready drafts produced without manual handoffs between steps; this workflow never publishes.

knuckles-team Bundle

End-to-end deployment of an infrastructure platform using docker-compose files and Portainer MCP.

knuckles-team Bundle

Full Docker infrastructure health assessment. Lists containers, images, volumes, and networks, then retrieves logs from a running container.

knuckles-team Bundle

Roll operating-system updates across a Kubernetes cluster one node at a time: discover healthy targets, process workers before control-plane nodes, cordon, drain, patch through an AgentConfig-resolved remote connection, reboot only when required, verify readiness, and uncordon. Use for governed node OS or kernel maintenance where workload availability must be preserved. Do not use for application rollouts, initial cluster bootstrap, or uncoordinated fleet-wide patching.

knuckles-team Bundle

Parallel execution workflow for feature development pipeline using the Unified Parallel Engine

knuckles-team Bundle

Deploys Prometheus, Grafana, and Loki in parallel and synthesizes dashboard integrations.

golemcloud Skill

Configuring the deployment logical version in the Golem Application Manifest (golem.yaml): the version: source (git tag, git commit hash, static string, or env var), tuning tagPattern/hashFallback/staticFallback/allowDirty, per-environment version overrides, and the versionCheck uniqueness policy. Use when setting up or changing how deploy versions are computed, comparing environments by version, or fixing version-related deploy errors.

golemcloud Skill

Setting up a Golem Cloud account from scratch. Use when creating a Golem Cloud account, authenticating with Golem Cloud via the CLI, setting up a cloud profile, or deploying to Golem Cloud for the first time.

golemcloud Skill

Configuring HTTP API domain deployments and security schemes in golem.yaml. Use when the user asks to deploy agents to a domain, configure API domains, set up authentication/security schemes (OIDC), or manage the httpApi section of the application manifest.

golemcloud Skill

Configuring MCP (Model Context Protocol) server deployments in golem.yaml. Use when the user asks to expose agents through MCP, enable MCP for an agent, add MCP server support, set up MCP authentication, or configure MCP deployments for local or cloud environments.

ericrisco Bundle

Use when calling a fal.ai endpoint by id to generate image, audio, or video from JS/Python/curl: subscribe vs submit, queue states, ED25519 webhook signature verification, per-call cost, or migrating off @fal-ai/serverless-client. NOT which model or art direction (that is ai-media); NOT the same models on another platform (that is replicate).

ericrisco Bundle

Use when operating a live n8n instance programmatically — its REST API (`{host}/api/v1`, `X-N8N-API-KEY`) or the n8n-mcp server — to create, validate, test, activate, update, delete workflows and read executions on self-host or n8n.cloud. NOT designing the flow or its importable JSON (that is `automation-flows`), NOT another platform (`make`, `zapier`).

tomevault-io Bundle

Azure Terraform Patterns Skill

xjli360 Bundle

Create evidence-backed Amazon image and A+ briefs with AI while preserving the real product's shape, color, scale, included components, and policy compliance. Use for AI做亚马逊商品图、主图差异化、副图和A+规划、评论洞察转视觉、图片测款. Not for fabricating the product.

xjli360 Bundle

Turn verified product facts and third-party keyword exports into an auditable AI workflow for filtering queries, selecting ad candidates, drafting listing fields, and preserving evidence for every claim. Use for AI写Listing、关键词表清洗、竞品词过滤、广告词选择、标题五点和后台词草拟. Not for invented claims or direct publishing.

legendtkl Skill

Kubernetes: horizontal pod autoscaling

legendtkl Skill

Terraform: drift detection and reconciliation

legendtkl Skill

Terraform: remote state and locking

legendtkl Skill

Terraform: importing existing resources

legendtkl Skill

Terraform: reusable module authoring

legendtkl Skill

Terraform: provider and multi-region configuration

chuyentn Skill

Azure Machine Learning SDK v2 for Python

chuyentn Skill

Azure Cosmos DB SDK for Python