OneLinersCommand workbench
Guides
Operational guides

Complete procedures with a way back.

Searchable, versioned procedures with prerequisites, verification, concrete output, and rollback.

Guides
243
Favorites
0
243 launch guides · 2601 reviewed steps
6 of 243 guides

AI/ML · Databases & Data · Security

Build a read-only PostgreSQL analytics bot with reviewed SQL and chart output

Build a local-first analytics assistant that discovers only curated PostgreSQL views, asks OpenAI for a structured SELECT proposal, validates a conservative SQL subset, requires a human to approve the exact query, enforces a restricted database role and read-only transaction, limits time and rows, and emits local table or chart JSON.
Save to collectionCreate a collection in the sidebar first.
caution240 minadvanced13 stepsUbuntu Server

AI/ML · Services & Applications · Software Engineering

Build a production Microsoft Teams assistant with the Agents SDK

Build a tenant-isolated Microsoft Teams assistant on the current Microsoft 365 Agents SDK with CloudAdapter JWT authentication, Microsoft Entra ID workload identity, durable activity idempotency, bounded read-only answers, Adaptive Card Action.Execute approvals, distinct app and user authorization, explicit admin-consent and resource-specific consent review, a separate dispatcher, and zero-downtime deployment.
Save to collectionCreate a collection in the sidebar first.
caution360 minadvanced14 stepsNode.js

AI/ML · DevOps & CI/CD · Software Engineering

Build a GitHub issue and pull-request triage bot with reviewed label writes

Create a production-oriented GitHub App that verifies webhook signatures, deduplicates deliveries, obtains short-lived installation authentication, proposes labels and duplicate candidates, and requires a named human to approve every external label write.
Save to collectionCreate a collection in the sidebar first.
caution220 minadvanced15 stepsUbuntu Server

AI/ML · Services & Applications · Security

Build a Gmail email triage and reply-draft bot with human approval

Build a single-mailbox production reference that incrementally reads Gmail, classifies untrusted messages, treats attachments as quarantined metadata, prepares bounded reply proposals, and creates—but never sends—a Gmail draft only after an identified reviewer approves the exact proposal.
Save to collectionCreate a collection in the sidebar first.
caution260 minadvanced14 stepsGmail API

AI/ML · Software Engineering · Security

Build a Claude tool-use support agent with approval gates

Build a support agent that uses strict Claude tool schemas, validates every tool request again in application code, executes independent read-only calls safely, pauses writes for human approval, returns correctly ordered tool results, and recovers from refusals, timeouts, and partial failures.
Save to collectionCreate a collection in the sidebar first.
caution210 minadvanced10 stepsNode.js

AI/ML · Software Engineering · Security · DevOps & CI/CD

Build a tool-using agent with human approvals, timeouts, retries, idempotency, and audit logs

Build a bounded Responses API agent whose strict function calls are treated as proposals, whose deterministic state and policy select eligible tools, whose consequential actions require exact human approval, and whose adapters use durable idempotency, effect-aware retries, deadlines, safe outputs, correlated audit and full-bundle rollback.
Save to collectionCreate a collection in the sidebar first.
caution300 minadvanced10 stepsOpenAI Responses API