tool-use-verification

Verify tool invocations with pre-flight and post-flight checks.

15|5|Updated May 17, 2026
One-click install
npx skills add https://github.com/cropsgg/skills --skill tool-use-verification
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: tool-use-verification
Source: https://github.com/cropsgg/skills/tree/main/skills/engineering/tool-use-verification
Command: npx skills add https://github.com/cropsgg/skills --skill tool-use-verification

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Pre-flight and post-flight verification ensures every tool invocation is scoped, safe, and auditable, preventing unintended state changes and enabling rapid diagnosis after executions.

Core Features & Use Cases

  • Pre-flight checks: Validate scope, safety, idempotency, and expected outcomes before every tool call.
  • Post-flight checks: Verify outputs, side effects, and error handling after execution.
  • Audit and recovery: Log decisions and results, enabling deterministic debugging and escalation when needed.

Quick Start

Always perform the 4 pre-flight questions before a tool call and the 4 post-flight checks after.

Frequently Asked Questions about tool-use-verification

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate pre-flight checks for shell commands and file edits?

Pre-flight checks for shell commands and file edits are automated by evaluating scope, safety, idempotency, and expected outcomes through a structured 4-question validation process before execution. This ensures deterministic and safe state mutations.

What is post-flight verification for tool invocations?

Post-flight verification for tool invocations is a structured 4-question assessment performed after execution to validate outputs, verify side effects, and handle errors. It enables rapid diagnosis and audit of state changes.

How do I ensure idempotency and safety for git operations?

Idempotency and safety for git operations are ensured by applying pre-flight and post-flight verification checks. The system validates scope, checks expected outcomes, and escalates on failures when guard mode is enabled.

Does this tool verification approach work without external dependencies?

Yes, this tool verification approach works without external dependencies. It applies to any environment where tools mutate state, implementing internal guardrails activated when guard mode is enabled to audit and recover from failures.

What's the best way to audit side effects after executing state-mutating tools?

The best way to audit side effects after executing state-mutating tools is using structured post-flight checks. The system logs decisions, verifies outputs, and records side effects to enable deterministic debugging and rapid recovery.