Index of every page on this site: llms.txt
All skills
rominirani avatar

/always-verify-gcp

@731acca
Run it yourself

A workflow skill for all Google Cloud resource management tasks (create, delete, modify). This skill ensures that the correct CLI tool (`gcloud`, `bq`, `gsutil`) is chosen and that its commands are verified against the latest official documentation before execution.

Use this Skill: https://skilld.dev/gh/rominirani/antigravity-skills/always-verify-gcp

Nothing lands on disk. Nothing to clean up.

Fork this Skill

Edit a local copy. It keeps the author and licence.

SKILL.md

≈71 tokens for metadata: the name and description. ≈725 when used: this file. ≈2.2k more on demand in 3 files.

Description uses 3.6% of example budget

Before choosing a Skill, your Agent reads its name and description. All available Skills share that space.

  • A shorter description leaves more room for other Skills. This entry exceeds our 1% size suggestion.

In our Claude Code example, all Skill names and descriptions share 8,000 characters. This Skill uses ≈286 characters, or 3.6%.

The 1% threshold is a size suggestion. Longer descriptions can still fit.

Your model, settings, and other Skills decide how much text your Agent can read.

Example settings and source

The example uses a 200k-token context and default Claude Code settings. The count includes the name, description, separators, and when_to_use when present. Codex also counts local file paths.

Skit's source and limits: Codex 0.160.1, Claude Code 2.1.292.

Always Verify GCP Skill (V3)

Overview

This skill establishes a rigorous, multi-step process for handling all Google Cloud resource management requests. Its goal is to maximize safety and accuracy by ensuring the correct tool is used, its commands are verified, its impact is previewed, and common errors are diagnosed intelligently.

Workflow: Verifying and Executing Google Cloud Commands

When a user requests an action involving Google Cloud resources, follow these steps:

Step 1: Select the Correct CLI Tool

Consult the references/tool_selection_guide.md file to determine the most appropriate command-line tool (gcloud, bq, or gsutil) based on the GCP service the user is targeting.

Step 2: Identify the Specific Command

Once the tool is selected, identify the specific command required to fulfill the user's request (e.g., bq mk, gsutil mb, gcloud compute instances delete).

Step 3: Verify Command with Official Documentation

  1. Consult MCP Server: Use search_documents to find the official documentation for the identified command. This is our connection to the "Developer Knowledge MCP Server" and is the primary source of truth.
  2. Use --help for Precision: If documentation is ambiguous, run the command with --help to get locally-relevant syntax and flags.

Step 4: Gather Missing Information

Using the verified parameters from the documentation, compare against the user's request. If required information is missing, use the ask_user tool to obtain it.

Step 5: Formulate the Final Command

Assemble the complete and accurate command.

Step 6: Perform a Dry Run (If Available)

  1. Check for Support: Before execution, check if the command supports a "dry run" or "preview" mode (e.g., gcloud ... --dry-run).
  2. Execute Dry Run: If supported, execute the command with the dry run flag first.
  3. Present Impact: Show the output of the dry run to the user, clearly explaining the changes that will be made (e.g., "This action will delete the following 3 instances: ...").
  4. Get Final Approval: Ask for a second, explicit approval to proceed with the actual execution.

If a dry run is not supported, explain this to the user and proceed to the next step.

Step 7: Propose and Execute

  1. Propose Command: If a dry run was not possible, present the final command to the user for their approval.
  2. Execute: Upon receiving user approval (either from this step or the dry run step), execute the command using run_shell_command.
  3. Report Outcome: Inform the user of the success or failure of the command.

Step 8: Diagnose Errors Intelligently

If the command fails, do not simply report the error. Instead, consult the references/error_handling_playbook.md file. Follow its instructions to diagnose the error and, if possible, propose a specific solution to the user.

Source: SKILL.md on GitHub

No rule matched.

skilld matched fixed text patterns in SKILL.md and file names. Patterns miss obfuscated code.

skilld run checks every file with the same patterns. It asks for approval before it loads a Skill with a behavior marked Needs approval.

No alerts2mo3 checks · Risk SAFE
  • Gen Agent Trust Hub2mo

    The skill provides a secure and methodical workflow for managing Google Cloud Platform resources, incorporating mandatory command verification and dry-run procedures.

  • Socket2mo

    No alerts

  • Snyk2mo

    Risk: LOW · No issues

Signed by skilld at 731acca. This ties the file your Agent reads to that commit on GitHub. It does not review the instructions.

Last checked against GitHub last month.

Steadyupdated 9 months ago

README badge

README badge for rominirani/antigravity-skills/always-verify-gcp