AGENT SKILL

verify-this

Verify a claim with fresh local evidence: restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE.

License 许可
MIT
最近更新
2026-11-15
Stars
★ 5
Forks
0

01 前言 Front Matter

name
verify-this
description
Verify a claim with fresh local evidence: restate it falsifiably, capture baseline and treatment, compare artifacts, and return VERIFIED, NOT VERIFIED, or INCONCLUSIVE.

03 SKILL.md 正文

Verify This

Verification is not a recap. It proves or disproves a specific claim with repeatable evidence.

When To Use

  • The user asks "verify this", "prove it works", "did this fix it", or "show me the evidence".
  • A bug fix needs a before/after repro.
  • A UI, CLI, API, performance, or memory claim needs measurement.
  • A test passes but the user-visible behavior still needs confirmation.

Do not use this for vague claims like "the code is cleaner". Ask for a measurable claim first.

Workflow

  1. Restate the claim in falsifiable form: condition, metric, and threshold.
  2. Pick the smallest local surface that can disprove it.
  3. Capture a baseline from the old state: merge base, parent commit, failing branch, or current broken repro.
  4. Capture treatment from the changed state with the same command, data, warmup, and environment.
  5. Compare raw artifacts: numbers, screenshots, terminal transcripts, HTTP responses, profiles, heap snapshots, or test output.
  6. Return exactly one verdict: VERIFIED, NOT VERIFIED, or INCONCLUSIVE.

Local Surfaces

  • Code behavior: focused unit/integration tests or a minimal repro script.
  • CLI/TUI behavior: control-cli, terminal transcript, or demo recording.
  • UI behavior: control-ui, screenshots, accessibility snapshots, or browser traces.
  • API behavior: local HTTP/RPC request and response diff.
  • Performance: same-machine baseline/treatment timings or CPU profiles.
  • Memory: heap snapshots before and after the suspected operation.

Artifact Layout

When safe to write artifacts:

/tmp/verify-this/<claim-slug>/
├── claim.md
├── timeline.md
├── baseline/
├── treatment/
├── diff/
└── verdict.md

If artifacts may contain sensitive code, prompts, screenshots, HTTP bodies, or heap data, keep only the minimal inline evidence unless the user agrees to disk storage.

Verdict Rules

  • VERIFIED: baseline and treatment differ in the predicted direction, by the claimed threshold, with no obvious confound.
  • NOT VERIFIED: the behavior is unchanged, moves the wrong way, or misses the threshold.
  • INCONCLUSIVE: no valid baseline, noisy signal, failed measurement, or an environment difference invalidates the comparison.

Output

Use this shape:

VERIFIED | NOT VERIFIED | INCONCLUSIVE
Claim: <falsifiable claim>

Evidence:
<metric/artifact>: baseline=<...>, treatment=<...>, delta=<...>, threshold=<...>

Reasoning:
<one tight paragraph naming the evidence and any confounds>

Do not soften a negative result. A clear NOT VERIFIED is useful.

04 同仓库更多 Skills

just-joshn
just-joshn/pi-customizations

plant

暂无简介

★ 5
just-joshn
just-joshn/pi-customizations

vitest-best-practices

Use when writing or reviewing Vitest tests, fixing flaky or slow tests, debugging failures, structuring suites, writing assertions (toBe vs toEqual vs toStrictEqual, asymmetric matchers, expect.soft, expect.poll, vi.waitFor), testing async code, setup/teardown hooks, test context and fixtures (test.extend, aroundEach), mocking (vi.fn, vi.spyOn, vi.mock, vi.hoisted, vi.when, classes), fake timers and dates, globals and env stubs, network mocking with MSW, file system mocking, snapshots, type testing (expectTypeOf), DOM environments (jsdom, happy-dom), Browser Mode, component testing (React, Vue, Svelte), benchmarking, database/integration tests, test tags and filtering, configuration and projects (monorepos), parallelism and performance, coverage, reporters, CI sharding, or migrating from Jest or Vitest 4. Targets Vitest 5.x.

★ 5
just-joshn
just-joshn/pi-customizations

plant-run

A planted skill whose blocks compile; one test fails, one is compile-only, and one never runs.

★ 5
just-joshn
just-joshn/pi-customizations

nope

A planted skill whose every file holds a mistake the gate must report, one class of mistake per block.

★ 5
just-joshn
just-joshn/pi-customizations

compose-robolectric-best-practices

Use when writing or reviewing Jetpack Compose UI tests that run on the JVM with Robolectric, fixing flaky or slow Compose tests, debugging test failures, structuring suites, choosing finders and semantics matchers (onNodeWithText, onNodeWithTag, hasX, useUnmergedTree), assertions, actions and gestures (performClick, performTextInput, performTouchInput, performScrollTo), synchronization (waitForIdle, waitUntil, mainClock, the v2 StandardTestDispatcher test APIs), JUnit rules and fixtures, screen robots, fakes and test data, ViewModel and StateFlow screens, navigation (Navigation Compose, Navigation 3), lists and scrolling, text fields and forms, dialogs and popups, animations and the test clock, configuration changes (qualifiers, dark mode, locale, RTL, font scale, state restoration), View interop, activities, intents, permissions and back press, screenshot tests with Roborazzi, accessibility checks, Robolectric configuration (@Config, sdk, qualifiers, looper and graphics modes), CI, performance, coverage, or migrating from the v1 Compose test APIs. Targets Compose BOM 2026.09 (ui-test 1.12) and Robolectric 4.17.

★ 5
just-joshn
just-joshn/pi-customizations

create-skill

Create Cursor Agent Skills. Use when authoring a new skill or asking about SKILL.md structure.

★ 5