verify

Verifies task completion by generating evidence against explicit Done criteria.

3|1|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/epologee/leclause-skills --skill verify-epologee
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: verify
Source: https://github.com/epologee/leclause-skills/tree/main/packages/autonomous/skills/verify
Command: npx skills add https://github.com/epologee/leclause-skills --skill verify-epologee

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The rover operates without direct supervision, so progress claims must be anchored to concrete, auditable evidence. This skill defines Done criteria up front and systematically gathers evidence to prove progress and completion.

Core Features & Use Cases

  • Define explicit Done criteria before work begins.
  • Verify progress by running code, capturing screenshots, curling endpoints, and inspecting state.
  • Invokable directly as /autonomous:verify on any loop file or used by the rover during SURVEY and INSPECT to validate results.

Quick Start

Run /autonomous:verify --propose <loop-file> to generate Done criteria and start collecting evidence.

Frequently Asked Questions about verify

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

FAQPage Schema
How do I verify autonomous rover task completion without direct supervision?

Verifying autonomous rover tasks without supervision requires defining explicit Done criteria up front and systematically collecting concrete evidence, such as endpoint responses or screenshots, to prove progress and completion.

What is evidence-based completion verification for autonomous systems?

Evidence-based completion verification is the process of anchoring progress claims to auditable data by pre-defining Done criteria and gathering verifiable proof during SURVEY and INSPECT rover operations.

How do I generate Done criteria from a loop file for rover inspection?

To generate Done criteria from a loop file, run the verify command with the propose flag on the target file, which automatically derives the criteria and begins collecting verification evidence.

Can I validate rover progress by curling endpoints and inspecting state?

Yes, you can validate rover progress by running code, capturing screenshots, curling endpoints, and inspecting system state to gather the verifiable evidence required to prove task completion.

Does this approach to rover verification require specific dependencies or components?

No, this verification approach requires no external dependencies or components, only the ability to read loop files, derive Done criteria, and produce verifiable evidence for each criterion.

When should I use predefined Done criteria for autonomous rover operations?

Predefined Done criteria should be used whenever a rover operates without direct supervision to ensure progress claims are systematically anchored to concrete, auditable evidence before work begins.