ralph

ralph is a skill for Claude Code from kangig94/coral. It costs 17 tokens per session (3,403 once invoked), scanned A, original, MIT.

A task-execution workflow that breaks work into steps and checks the result before declaring it complete.

In plain words
What is it for?
Use it to carry out coding tasks, verify acceptance criteria, run adversarial tests, or execute independent work in parallel.
Why use it?
It helps prevent unfinished work and unsupported completion claims by requiring fresh evidence such as test, build, or lint output.

Skill for Claude Code

Written for Claude Code: argument-hint in frontmatter. Also seen: mentions subagents; names the AskUserQuestion tool; mentions Codex.

Part of the coral plugin — 12 skills, 10 agents shipped together

Good fit Use it to carry out coding tasks, verify acceptance criteria, run adversarial tests, or execute independent work in parallel.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/kangig94/coral/ralph
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.

Any agent
npx skills add kangig94/coral --skill ralph
Clone the repo
git clone --depth 1 https://github.com/kangig94/coral

Made for: Claude Code.

Or install coral, the plugin that ships this one along with the rest of its 12 skills, 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 ralph

README.md
[![agentmods](https://agentmods.dev/badge/skills/kangig94/coral/ralph.svg)](https://agentmods.dev/skills/kangig94/coral/ralph)
Your own site
<a href="https://agentmods.dev/skills/kangig94/coral/ralph"><img src="https://agentmods.dev/badge/skills/kangig94/coral/ralph.svg" alt="Measured on agentmods" height="20"></a>
Per session 17 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,403 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. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
SkillSpector: 3 findings, up to medium

These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →

  • medium Prompt Injection · line 13
    Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
    Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
  • medium Prompt Injection · line 15
    Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.
    Fix: Remove the large whitespace padding (blank-line blocks or long space runs) and review any content hidden below or to the right of it. Keep skill files compact and reviewable so no instructions can be
  • medium Excessive Agency · line 75
    Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.
    Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
How audits are shown
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 $0.00017 $0.03403
Opus 5 $0.00009 $0.01702
Sonnet 5 $0.00003 $0.00681
Haiku 4.5 $0.00002 $0.00340

Measured 8d ago against content hash a61c79911f0c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

ralph 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 8d 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.

clients/skills/ralph/SKILL.md · 248 lines

How it starts

The opening of the file, as written. The whole thing — 248 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Persistent Execution with Verification

Announce at start: "Using ralph to execute this task with verification loop."

Argument Routing

Argument Mode
<prompt> Self-execute on current host (default)
--delegate Delegate to the other host (Claude → Codex, Codex → Claude, Copilot → Codex; from SessionStart Current host:)
--red Adversarial testing (spawns red-attacker in parallel)
--team Parallel AC execution via Agent Teams (plan mode only)

Strip flags before passing the prompt to execution. Preserve original flags in the state file prompt for resume continuity.

<Ralph_Protocol> You are Ralph — a persistent task executor. Complete tasks fully with verified evidence. Responsible for: breaking tasks into steps, executing, verifying completion with evidence. Not responsible for: requirements (gap-finder), plan review (critic), architecture (architect). Parallelize independent work — never wait sequentially for independent tasks. <Success_Criteria> NO COMPLETION CLAIMS WITHOUT FRESH VERIFICATION EVIDENCE

- Every completion claim is backed by fresh verification output (test/build/lint)
- All acceptance criteria from the original task are met (no scope reduction)
- Post-implementation sequence passes in order: lint → validation → build → test

</Success_Criteria>

DO DON'T
Implement every AC fully as written Stub, skeleton, placeholder, or partial implementation
Pass AC text verbatim to every delegate Rephrase, simplify, defer, or omit any part of an AC
Treat AC complexity as the job, not an obstacle Judge an AC as "too complex" and reduce its scope
Run build/test only in post-implementation Run build or test during implementation
Verify subagent output independently Trust "agent said success"
Escalate to architect after 3 failed fix attempts Try variations of the same fix
Output <promise> only after ALL verification passes Output false promise to escape the loop

Read the full file on GitHub · 248 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. 8d ago First seen · 248 lines · 17 tokens per session scan A a61c79911f0c

Subscribe to this mod's changes

ralph is a skill published in the GitHub repository kangig94/coral (11 stars, last pushed 5d ago), licensed MIT. It adds 17 tokens to every session and 3,403 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 skills, from other repositories

taiyi-dev

A software-development stage that implements planned tasks using test-driven development, or TDD: write a failing test, make it pass, then improve the code. It checks the task plan, dependencies, file boundaries, and required completion evidence.

Dong90/oh-my-taiyiforge · 21 tokens

cherry-pr-test

Test Cherry Studio PRs by resolving and checking out a PR, statically inspecting its changes, running interactive UI tests against a safely tracked Electron instance through CDP, producing a structured report, cleaning up only the owned test instance, and restoring the original branch.

CherryHQ/cherry-studio · 57 tokens

taiyi-test

A project workflow skill for verifying an implementation and producing a TEST.md record. TDD means writing a failing test, implementing the change, and then making the test pass; this skill checks that process and other regression cases.

Dong90/oh-my-taiyiforge · 24 tokens

flow-next-qa

Live-app QA pass derived from the spec. Drives the running app, files P0/P1/P2 findings with evidence, emits a YES or NO qaverdict receipt.

gmickel/flow-next · 40 tokens

flow-next-features

Seed or maintain the committed user-POV drive map at .flow/features/ so QA and drive reuse how a user reaches each feature.

gmickel/flow-next · 33 tokens

memstack-development-test-writer

Use this skill when the user says 'write tests', 'add tests', 'test coverage', 'unit tests', 'integration tests', 'component tests', 'mocking', 'edge cases', or needs to generate tests with proper mocking and edge case coverage. Do NOT use for refactoring plans or database migrations.

cwinvestments/memstack · 70 tokens