tinybird-cli-guidelines

Overview of tinybird-cli-guidelines skill

The tinybird-cli-guidelines skill is a practical guide for using the Tinybird CLI (tb) with fewer guesswork errors and better workflow decisions. It is best for backend developers, data engineers, and agents that need to build, validate, deploy, or operate Tinybird projects without inventing flags or running the wrong target by accident.

What users usually care about is simple: “Which tb command should I run, in which environment, and what should I check before I change data or deploy?” This skill focuses on that job-to-be-done. The tinybird-cli-guidelines guide is strongest when you need local development, Cloud deployment, data operations, mock data, or token/secret handling to stay consistent.

Why this skill is worth installing

The main value is workflow control, not command discovery alone. tinybird-cli-guidelines install is useful if you want a skill that emphasizes CLI 4.0 behavior, environment targeting, and safe operational habits. It helps reduce errors around tb build, tb deploy, destructive data changes, and local-vs-cloud confusion.

Best fit for backend development

Use tinybird-cli-guidelines for Backend Development when your work touches Tinybird datasources, pipes, deployments, or local testing. It is a good fit if you need a compact operating guide for the tb CLI, especially in repositories where build/deploy behavior depends on tinybird.config.json.

What it does not try to do

This is not a generic Tinybird tutorial or a full repo summary. It is a decision-support skill for CLI usage and operational flow. If you only need a one-off command explanation, tb <command> --help may be enough; if you need a repeatable workflow, the skill adds more value.

How to Use tinybird-cli-guidelines skill

Install it and verify the skill scope

Install tinybird-cli-guidelines with:

npx skills add tinybirdco/tinybird-agent-skills --skill tinybird-cli-guidelines

After install, read SKILL.md first, then open the supporting rule files that match your task. The most useful files are rules/cli-commands.md, rules/build-deploy.md, rules/local-development.md, and rules/data-operations.md.

Turn a rough request into a usable prompt

The skill works best when your prompt includes:

• the goal: build, deploy, append, replace, delete, or test
• the environment: local, branch, or cloud
• the asset: datasource, pipe, token, secret, or project file
• the risk level: safe check, non-destructive, or confirmed destructive change

A stronger prompt looks like: “Use tinybird-cli-guidelines to validate a Tinybird project, explain which tb commands to run in CLI 4.0, and avoid destructive deploys unless I confirm.”

Read the right files first

Start with SKILL.md, then use the rule file that matches the task:

• rules/build-deploy.md for tb build and tb deploy
• rules/data-operations.md for delete/replace/truncate behavior
• rules/append-data.md for appending files, URLs, or events
• rules/local-development.md for Tinybird Local setup and troubleshooting
• rules/tokens.md and rules/secrets.md for auth and configuration work

Use the workflow rules, not memory

The highest-value guidance in tinybird-cli-guidelines usage is to respect CLI 4.0 context. Configure dev_mode in tinybird.config.json, then use plain tb build and tb deploy unless you truly need manual overrides. Before any command that could change production or delete data, check the exact command syntax in the skill or confirm it with tb <command> --help.

tinybird-cli-guidelines skill FAQ

Is tinybird-cli-guidelines only for Tinybird experts?

No. It is useful for beginners who need safer defaults and for experienced users who want fewer environment mistakes. The skill is especially helpful when you know your business goal but not the safest Tinybird CLI path.

How is this different from a normal prompt?

A normal prompt can describe the task, but tinybird-cli-guidelines gives you a reusable operating model: where to look, what to confirm, and when to stop and ask for approval. That matters most for tb build, tb deploy, and data mutation commands.

When should I not use this skill?

Skip it if you are not working with Tinybird CLI commands, Tinybird Local, or Tinybird deployment workflows. It is also unnecessary for very simple read-only questions that do not involve project files, environments, or operational risk.

Does it cover both local and cloud workflows?

Yes. The tinybird-cli-guidelines guide covers Tinybird Local, branch-based development, and Cloud deployment. That makes it a good fit when you need to move from local iteration to production with fewer manual decisions.

How to Improve tinybird-cli-guidelines skill

Give the skill the exact operation and target

The better your input, the better the output. Instead of “help me with Tinybird,” say “run a safe local build,” “prepare a production deploy check,” or “append CSV data to an existing datasource.” Include the datasource or project name, and say whether the target is local, branch, or cloud.

State your constraints up front

The biggest failure mode is unclear risk. If you cannot allow destructive changes, say so before the skill suggests tb deploy --allow-destructive-operations or a replacement command. If you are working in main or master, mention that too, because branch-mode behavior is different.

Provide concrete inputs for data work

For tinybird-cli-guidelines usage on append, replace, or delete tasks, share the file path, row condition, partition key, and whether the schema is unchanged. For example, “Replace November rows in events using data.csv; partition key is country; do not touch active ingestion.” That level of detail avoids unsafe or incomplete command plans.

Iterate with checks before production

Use tb info, tb deploy --check, and the relevant rule file before you promote anything to Cloud main. If the first result is too generic, refine the request by adding the command, the environment, and the exact files you want the skill to inspect.