What is the difference between verification and sniper?

Verification validates functional resolution, while sniper validates code quality and style.

What counts as evidence?

Test outputs, logs, UI screenshots, diffs, and any authoritative artifact showing the behavior meets acceptance criteria.

What happens after verification passes?

If both verification and sniper pass, the task is closed; otherwise it remains open until issues are resolved.

verification

npx machina-cli add skill fusengine/agents/verification --openclaw

Files (1)

SKILL.md

3.2 KB

Verification Before Completion

Overview

Sniper validates CODE QUALITY. Verification validates FUNCTIONAL RESOLUTION. Both are needed before closing any task.

Sniper catches linter errors, SOLID violations, and code style issues. Verification ensures the original request is actually fulfilled -- the right behavior, the right output, the right fix. A task can pass sniper with zero errors and still be functionally wrong.

Aspect	Sniper	Verification
Focus	Code quality	Functional correctness
Checks	Linting, SOLID, style	Acceptance criteria, regressions, side effects
Runs	After any code change	Before marking task as complete
Result	Clean code	Solved problem

Agent Workflow

When invoked, follow the 6-step verification process below before marking any task as completed.

6-Step Verification Process

Step 1: Re-read the original request Go back to the original issue, task description, or user message. Read it word by word. Do not rely on memory or assumptions.

Step 2: List ALL acceptance criteria Extract every explicit and implicit requirement from the original request. Number them. If the request is vague, list what a reasonable user would expect.

Step 3: Verify each criterion with evidence For each criterion, provide concrete evidence of resolution:

Test output showing the expected behavior
Log output confirming the fix
Screenshot of the UI change
Code diff showing the implementation

Step 4: Check for regressions Run the full test suite. Compare results before and after. No new failures, no new warnings.

Step 5: Check for side effects Review every modified file. Confirm no accidental changes to unrelated code. Verify dependencies and configuration are unchanged unless required.

Step 6: Confirm functional resolution State explicitly: "Original problem is FUNCTIONALLY resolved" with a summary of evidence, or list what remains unresolved.

Reference Guide

Resource	Path	Content
Checklist	`references/checklist.md`	Full verification checklist with all categories
Common Misses	`references/common-misses.md`	Frequently forgotten verification items

Integration with APEX

Verification runs between eLicit and eXamine in the APEX workflow:

Analyze -> Plan -> Execute -> eLicit -> [VERIFICATION] -> eXamine (sniper)

This ensures functional correctness is confirmed before code quality validation. A task is only complete when both verification and sniper pass.

Critical Rules

Rule	Reason
Never skip re-reading the original request	Prevents solving the wrong problem
Evidence required for every criterion	"It works" is not evidence
Full test suite, not just new tests	Catches regressions
Review ALL modified files	Catches accidental side effects
Both verification AND sniper must pass	Quality without correctness is useless

Source

git clone https://github.com/fusengine/agents/blob/main/plugins/ai-pilot/skills/verification/SKILL.mdView on GitHub

Overview

Verification confirms the requested functionality is actually fulfilled before closing a task. It complements code quality checks by validating acceptance criteria, regressions, and potential side effects.

How This Skill Works

Follow the 6-step verification process: reread the original request, list all acceptance criteria, verify each with concrete evidence, check for regressions, review side effects, and confirm functional resolution. In the APEX workflow, verification runs between eLicit and eXamine to ensure a task is only closed after both verification and code quality checks pass.

When to Use It

After implementing a feature, to mark the task complete only if it meets acceptance criteria.
When a bug fix is ready, to verify the original issue is resolved with evidence.
Before closing a user story, to confirm the requested behavior is fulfilled.
After code changes that could affect related functionality, to check regressions.
When side effects or integration points were modified, to verify no unintended changes.

Quick Start

Step 1: Re-read the original request
Step 2: List ALL acceptance criteria with numbers
Step 3: Attach evidence for each criterion and declare FUNCTIONALLY RESOLVED if satisfied

Best Practices

Reread the exact original request word by word
Enumerate all explicit and implicit acceptance criteria
Attach concrete evidence for every criterion (tests, logs, screenshots, diffs)
Run the full test suite and compare results
Ensure both Verification and Sniper pass before closing

Example Use Cases

Bug fix: login flow now redirects correctly; attach test output and a diff of the fix
Feature: export summarizes data as expected; provide UI screenshot and sample output
API change: pagination works as documented; include API response diff and test results
Data migration: data integrity maintained post-migration; attach verification logs
UI change: input validation now matches criteria; include automated test results

Frequently Asked Questions

Add this skill to your agents