How do I install the Redsub Validate skill?

Run npx skillmds@latest add tomevault-io/redsub-validate in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

What does the Redsub Validate skill do?

Run lint, type check, and unit tests with evidence. Use when this capability is needed. It is listed under Coding & Dev Tools on SkillMD.

Is Redsub Validate safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Redsub Validate?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Redsub Validate free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Redsub Validate?

tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Redsub Validate by tomevault-io — Agent Skill

Name: Redsub Validate
Author: tomevault-io

File contents

Code Validation

Command Resolution

Determine the project's validation commands:

Check project CLAUDE.md for explicit commands (lint, check, test)
If not found, read package.json scripts and infer:
- Lint: script containing "lint" (e.g., lint, lint:fix)
- Type check: script containing "check" (e.g., check, typecheck)
- Unit test: script containing "test" (e.g., test, test:unit)
Detect package manager: look for lock files (pnpm-lock.yaml → pnpm, yarn.lock → yarn, default → npm)
If ambiguous, ask the user

Procedure

Run sequentially. Stop on first failure.

1. Lint

<package-manager> run <lint-script>

2. Type check

<package-manager> run <check-script>

3. Unit tests

<package-manager> run <test-script>

If the test script doesn't include a --run flag for watch-mode frameworks (e.g., vitest), append -- --run.

4. SSOT Consistency Checks

When a canonical source exists (config, constants, types), verify consumers stay in sync:

Config-driven values: verify runtime reads match the canonical source (e.g., routes file exports match router config)
Shared types: verify API response shapes match the declared type (snapshot or schema validation)
i18n: verify all keys used in components exist in translation files
If a test duplicates a magic number or string, extract to shared fixture or import from source module

5. Evidence (5-Step Verification Gate)

No claims without evidence. Follow this exact sequence:

IDENTIFY: What command proves this claim?
RUN: Execute the full command (fresh, complete)
READ: Check entire output, exit code, failure count
VERIFY: Does the output support the claim?
CLAIM: Only then make the claim

Forbidden before verification: "should", "probably", "seems to", or satisfaction expressions ("Great!", "Done!", "Perfect!").

On success:

Validation passed (with evidence):
- lint: pass (0 errors, 0 warnings)
- type check: pass (0 errors)
- unit test: pass (N tests, 0 failures)

On failure:

Validation failed:
- [step]: [error details]
- Fix: [suggestion]

Converted and distributed by TomeVault — claim your Tome and manage your conversions.

---name: redsub-validatedescription: Run lint, type check, and unit tests with evidence. Use when this capability is needed.---# Code Validation## Command ResolutionDetermine the project's validation commands:1. Check project CLAUDE.md for explicit commands (lint, check, test)2. If not found, read `package.json` scripts and infer: - Lint: script containing "lint" (e.g., `lint`, `lint:fix`) - Type check: script containing "check" (e.g., `check`, `typecheck`) - Unit test: script containing "test" (e.g., `test`, `test:unit`)3. Detect package manager: look for lock files (`pnpm-lock.yaml` → pnpm, `yarn.lock` → yarn, default → npm)4. If ambiguous, ask the user## ProcedureRun sequentially. Stop on first failure.### 1. Lint```bash<package-manager> run <lint-script>```### 2. Type check```bash<package-manager> run <check-script>```### 3. Unit tests```bash<package-manager> run <test-script>```If the test script doesn't include a `--run` flag for watch-mode frameworks (e.g., vitest), append `-- --run`.### 4. SSOT Consistency ChecksWhen a canonical source exists (config, constants, types), verify consumers stay in sync:- Config-driven values: verify runtime reads match the canonical source (e.g., routes file exports match router config)- Shared types: verify API response shapes match the declared type (snapshot or schema validation)- i18n: verify all keys used in components exist in translation files- If a test duplicates a magic number or string, extract to shared fixture or import from source module### 5. Evidence (5-Step Verification Gate)**No claims without evidence.** Follow this exact sequence:1. **IDENTIFY**: What command proves this claim?2. **RUN**: Execute the full command (fresh, complete)3. **READ**: Check entire output, exit code, failure count4. **VERIFY**: Does the output support the claim?5. **CLAIM**: Only then make the claimForbidden before verification: "should", "probably", "seems to", or satisfaction expressions ("Great!", "Done!", "Perfect!").**On success:**```Validation passed (with evidence):- lint: pass (0 errors, 0 warnings)- type check: pass (0 errors)- unit test: pass (N tests, 0 failures)```**On failure:**```Validation failed:- [step]: [error details]- Fix: [suggestion]```---> Converted and distributed by [TomeVault](https://tomevault.io/claim/redsub-captain) — claim your Tome and manage your conversions.

Frequently asked questions

Redsub Validate