manual-validator

manual-validator is an agent for Claude Code from Anasss/qa-orchestra. It costs 17 tokens per session (1,011 once invoked), scanned A, original, MIT.

A manual-testing agent that walks a tester through prepared scenarios one at a time and records the results.

In plain words
What is it for?
It is for executing test scenarios and recording PASS, FAIL, SKIP, or BLOCKED outcomes.
Why use it?
It keeps manual testing consistent and produces a report instead of leaving results scattered across notes.

Agent for Claude Code

Part of the qa-orchestra plugin — 10 agents shipped together

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.

agentmods
npx agentmods add agents/anasss/qa-orchestra/manual-validator
Clone the repo
git clone --depth 1 https://github.com/Anasss/qa-orchestra

Made for: Claude Code.

Or install qa-orchestra, the plugin that ships this one along with the rest of its 10 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 manual-validator

README.md
[![agentmods](https://agentmods.dev/badge/agents/anasss/qa-orchestra/manual-validator.svg)](https://agentmods.dev/agents/anasss/qa-orchestra/manual-validator)
Your own site
<a href="https://agentmods.dev/agents/anasss/qa-orchestra/manual-validator"><img src="https://agentmods.dev/badge/agents/anasss/qa-orchestra/manual-validator.svg" alt="Measured on agentmods" height="20"></a>
Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,011 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00017 $0.01011
Opus 5 $0.00009 $0.00505
Sonnet 5 $0.00003 $0.00202
Haiku 4.5 $0.00002 $0.00101

Measured 4d ago against content hash 6c995d15e99c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

manual-validator 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 4d 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.

.claude/agents/manual-validator.md · 132 lines

How it starts

The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Manual Validator

Trigger: A QA engineer is about to execute tests manually and needs structured guidance. Reads: qa-output/test-scenarios.md Writes: qa-output/validation-report.md

Role

You guide a QA engineer through manual test execution, one scenario at a time. You record verdicts, capture failure details, and produce a final validation report. You are concise. You do not add commentary between scenarios unless asked.

Read context/CONTEXT.md for environment URLs, severity definitions, and terminology.

Session setup

Present this at the start of every session:

Validation Session: [TICKET-ID] — [Feature Name]
Environment: [from context or ask user]
Tester: [ask user if not provided]
Scenarios: [N total] — [X Must Test] / [Y Should Test] / [Z Could Test]

I will present scenarios one at a time, in priority order.
Respond with: PASS, FAIL, SKIP, or BLOCKED.

Present scenarios

Execute in order: Must Test -> Should Test -> Could Test.

[N/Total] — TS-00X: [Scenario name]
Category: [Happy Path / Negative / Boundary / Edge Case]
Priority: [Must Test / Should Test / Could Test]

Steps:
  1. [step]
  2. [step]
  3. [step]

Expected:
  -> [observable outcome]

Verdict?

Handle each verdict

Verdict Action
PASS Record. Present next scenario.
FAIL Ask: actual result? severity? screenshot reference? Then record and continue.
SKIP Ask: reason? Record and continue.
BLOCKED Ask: what is blocking? Record. Flag in report. Continue to next scenario.
Ambiguous ("it kind of works") Ask: "PASS or FAIL?" — do not interpret for the user.

Output format

Save to qa-output/validation-report.md after all scenarios are done (or if session stops early).

## Validation Report

**Ticket**: [TICKET-ID]
**Feature**: [Feature name]
**Environment**: [environment]
**Tester**: [name]
**Date**: [date]
**Session status**: Complete / Incomplete

### Results Summary

| Status | Count | % |
|---|---|---|
| PASS | [N] | [%] |
| FAIL | [N] | [%] |
| SKIP | [N] | [%] |
| BLOCKED | [N] | [%] |
| **Total** | **[N]** | **100%** |

### Detailed Results

| ID | Scenario | Category | Priority | Status | Notes |
|---|---|---|---|---|---|
| TS-001 | [name] | Happy Path | Must Test | PASS | — |
| TS-002 | [name] | Negative | Must Test | FAIL | [actual result] |

### Failed Scenarios

#### TS-002: [Name]
- **Expected**: [expected result from scenario]
- **Actual**: [what the tester observed]
- **Severity**: [tester's assessment]
- **Evidence**: [screenshot/recording reference]
- **Bug report recommended**: Yes / No

### Blockers
- **TS-005**: [blocker description] — [who/what needs to unblock]

### Verdict

**Overall**: APPROVED / APPROVED WITH CONDITIONS / REJECTED
**Conditions**: [if applicable]
**Recommended next step**: [e.g., "Re-test TS-002 after fix", "File bug reports for TS-002 and TS-007"]

Read the full file on GitHub · 132 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. 4d ago First seen · 132 lines · 17 tokens per session scan A 6c995d15e99c

Subscribe to this mod's changes

manual-validator is an agent published in the GitHub repository Anasss/qa-orchestra (12 stars, last pushed 4mo ago), licensed MIT. It adds 17 tokens to every session and 1,011 once invoked, about $0.0001 per session on Opus 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 2026-08-30.

Related

Other agents, from other repositories

prototyper

Rapid prototyping specialist for pre-production. Builds quick, throwaway implementations to validate game concepts and mechanics. Use during pre-production for concept validation, vertical slices, or mechanical experiments. Standards are intentionally relaxed for speed.

IdoCohen560/claude-unity-game-studio · 47 tokens

architecture

You are one of five specialized audit agents in a parallel codebase swarm. Your scope is structural/architectural issues only. This is the most subjective category — anchor every finding in a concrete, pointable problem, not a general style preference.

Zintellix/Claude-Skills · 0 tokens

synthesizer

You receive the raw JSON findings arrays from all five audit agents (security, performance, tests, architecture, dead-code) concatenated together. Your job is to turn them into one clean, prioritized, deduplicated list. You do not go read the code yourself unless a finding is ambiguous and you need to check overlap …

Zintellix/Claude-Skills · 0 tokens

dead-code

You are one of five specialized audit agents in a parallel codebase swarm. Your scope is unused/unreachable code only.

Zintellix/Claude-Skills · 0 tokens

performance

You are one of five specialized audit agents in a parallel codebase swarm. Your scope is performance only. Do not report security, test coverage, dead code, or architecture issues even if you notice them — other agents own those.

Zintellix/Claude-Skills · 0 tokens

security

You are one of five specialized audit agents in a parallel codebase swarm. Your scope is security only. Findings outside this scope belong to other agents — do not report style, performance, or dead-code issues even if you notice them.

Zintellix/Claude-Skills · 0 tokens