Skill

verify

Evidence-based completion check. Run before claiming any task is done. Prevents the trust-then-verify gap. Use after implementing a feature, fix, or refactor.

Install

npx claudepluginhub artmin96/forge-studio --plugin evaluator

Tool Access

This skill is limited to using the following tools:

ReadBashGlobGrep

SKILL.md

Similar Skills

cache-components

Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.

cache-components

138.8k

claude-opus-4-5-migration

2 files

Migrates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings on Anthropic, AWS, GCP, Azure platforms.

claude-opus-4-5-migration

83.2k

bmad-brainstorming

11 files

Facilitates interactive brainstorming sessions using diverse creative techniques and ideation methods. Activates when users say 'help me brainstorm' or 'help me ideate'.

bmad-pro-skills

43.8k

Stats

Parent Repo Stars1

Parent Repo Forks0

Last CommitApr 3, 2026

Actions

View Source View Plugin View on GitHub View README

5. Verdict

VERIFIED: [Yes/No] METHOD: [tests/build/manual/type-check] EVIDENCE: [One line summary of proof] REMAINING RISK: [What could still go wrong, or "None identified"]

If you CANNOT verify the change:

UNVERIFIED: Cannot verify this change. NEEDED: [What would be needed to verify — test command, expected output, etc.]

Never claim work is done without evidence. Evidence, not assertions.

Verify: Evidence Before Assertions

From Anthropic's best practices: "The single highest-leverage thing you can do is include tests, screenshots, or expected outputs so Claude can verify its own work."

Before marking this task complete, answer EVERY question:

1. What Changed?

List every file modified and what changed in each (one line per file)
Run git diff --stat to confirm

2. What's the Verification Method?

Pick one or more:

Tests: Run the test suite. Show output. All pass? Which tests cover this change?
Build: Does it compile/build without errors? Run the build command.
Manual check: Describe what to look at and what the expected behavior is.
Type check: Run static analysis if available.

3. Run the Verification

Actually run it. Show the output. Don't say "it should work" — show that it DOES work.

3b. Ground in Actual Output

Quote the actual output — don't paraphrase it. Copy-paste the real test output, build output, or command result. If you can't quote real output, you haven't verified.

Bad: "Tests pass successfully." Good: "Output: Tests: 42 passed, 0 failed (0.83s)"

4. Edge Cases

What's the most likely way this breaks?
Did you handle null/empty/boundary inputs?
What happens if this runs twice?

5. Verdict

VERIFIED: [Yes/No]
METHOD: [tests/build/manual/type-check]
EVIDENCE: [One line summary of proof]
REMAINING RISK: [What could still go wrong, or "None identified"]

If you CANNOT verify the change:

UNVERIFIED: Cannot verify this change.
NEEDED: [What would be needed to verify — test command, expected output, etc.]

Never claim work is done without evidence. Evidence, not assertions.