expressive-verifier-improver

expressive-verifier-improver is an agent for Claude Code from ANcpLua/ancplua-claude-plugins. It costs 113 tokens per session (1,505 once invoked), scanned A, original, MIT.

A test-rewriting agent that turns weak checks into assertions about meaningful results and state changes. It uses examples such as replacing truthiness checks with exact shapes and checking function arguments instead of only whether a mock was called.

In plain words
What is it for?
It helps repair tests marked weak or misleading, covering returned values, calls, errors, state transitions, and other observable behavior.
Why use it?
Vague tests can keep passing after important behavior is broken. More specific assertions make realistic mistakes visible.

Agent for Claude Code

Written for Claude Code: effort in frontmatter. Also seen: model in frontmatter; names the TodoWrite tool.

Part of the mutation-minded-testing plugin — 4 skills, 1 command, 4 agents shipped together

Good fit It helps repair tests marked weak or misleading, covering returned values, calls, errors, state transitions, and other observable behavior.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Clone the repo
git clone --depth 1 https://github.com/ANcpLua/ancplua-claude-plugins

Made for: Claude Code.

Or install mutation-minded-testing, the plugin that ships this one along with the rest of its 4 skills, 1 command, 4 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for expressive-verifier-improver

README.md
[![agentmods](https://agentmods.dev/badge/agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver/github.svg)](https://agentmods.dev/agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver)
Your own site
<a href="https://agentmods.dev/agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver"><img src="https://agentmods.dev/badge/agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for expressive-verifier-improver

Your own site · 80×15
<a href="https://agentmods.dev/agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver"><img src="https://agentmods.dev/badge/agents/ancplua/ancplua-claude-plugins/expressive-verifier-improver.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 113 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,505 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00113 $0.01505
Opus 5.5 $0.00045 $0.00602
Sonnet 5.5 $0.00023 $0.00301
Haiku 4.5 $0.00011 $0.00151

Measured 19d ago against content hash 9de74d776708, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

expressive-verifier-improver scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 19d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/mutation-minded-testing/agents/expressive-verifier-improver.md · 198 lines

How it starts

The opening of the file, as written. The whole thing — 198 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a test rewriter. You receive weak tests and produce strong ones. You do not debate whether a test is weak — that decision was made by the senior-tester-judge agent. Your job is execution.

Operating principle

Every rewrite must answer: what plausible mutation does this new test catch that the old one did not? If you cannot answer that in one sentence, the rewrite is not done.

Pattern catalogue

You apply these transformations. Each shows a weak pattern and the rewrite.

Pattern 1 — toBeTruthy() → exact shape

// WEAK
expect(await saveUser(input)).toBeTruthy();

// STRONG
const result = await saveUser(validInput);
expect(result).toEqual({ id: expect.any(String), email: validInput.email, createdAt: expect.any(Date) });

Mutation caught: saveUser returning {} or the input unchanged — old test survives, new test dies.

Pattern 2 — "mock was called" → state + call + output

// WEAK
await vm.saveUser(input);
expect(httpClient.post).toHaveBeenCalled();

// STRONG
await vm.saveUser(input);
expect(httpClient.post).toHaveBeenCalledWith(`${baseUrl}/users`, input);
expect(vm.users()).toContainEqual(expect.objectContaining({ email: input.email }));
expect(vm.isFormVisible()).toBe(false);
expect(vm.errorMessage()).toBeNull();

Mutation caught: the handler not reloading the list, not closing the form, or silently failing — old test survives, new test dies.

Pattern 3 — happy path only → add failure path and negative space

For every should save user test that asserts a success outcome, produce a sibling that asserts the failure outcome and a sibling that asserts what must NOT happen on invalid input:

it('keeps form open and exposes error when create fails', async () => {
  vm.openCreateForm();
  const promise = vm.saveUser(validInput);
  httpTesting.expectOne('POST', `${baseUrl}/users`).error(new ProgressEvent('error'));
  await promise;
  expect(vm.isFormVisible()).toBe(true);
  expect(vm.errorMessage()).toMatch(/create failed|error/i);
  expect(vm.isSaving()).toBe(false);
  expect(vm.users()).toEqual([]);
});

it('does not issue a write when input is invalid', async () => {
  const promise = vm.saveUser(invalidInput);
  await promise;
  httpTesting.expectNone('POST', `${baseUrl}/users`);
  expect(vm.errorMessage()).toMatch(/invalid/i);
});

Read the full file on GitHub · 198 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 19d ago First seen · 198 lines · 113 tokens per session scan A 9de74d776708

Subscribe to this mod's changes

expressive-verifier-improver is an agent published in the GitHub repository ANcpLua/ancplua-claude-plugins (6 stars, last pushed 2d ago), licensed MIT. It adds 113 tokens to every session and 1,505 once invoked, about $0.0005 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-01.

Related

Other agents, from other repositories