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.
npx skills add ToruAI/toru-claude-agents --skill dev-full-autogit clone --depth 1 https://github.com/ToruAI/toru-claude-agentsWrote 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.
[](https://agentmods.dev/skills/toruai/toru-claude-agents/dev-full-auto)<a href="https://agentmods.dev/skills/toruai/toru-claude-agents/dev-full-auto"><img src="https://agentmods.dev/badge/skills/toruai/toru-claude-agents/dev-full-auto/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.
<a href="https://agentmods.dev/skills/toruai/toru-claude-agents/dev-full-auto"><img src="https://agentmods.dev/badge/skills/toruai/toru-claude-agents/dev-full-auto.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00030 | $0.02560 |
| Opus 5 | $0.00015 | $0.01280 |
| Sonnet 5 | $0.00006 | $0.00512 |
| Haiku 4.5 | $0.00003 | $0.00256 |
Grade A, and why
dev-full-auto scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -X POST "https://api.telegram.org/bot${TELEGRAM_BOT_TOKEN}/sendMessage" \ How it starts
The opening of the file, as written. The whole thing — 408 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Dev Full Auto - Autonomous Development
Spec in. Production-ready software out. I make decisions, you review results.
Philosophy: The Ralph Loop
"The best code review is reviewing working software, not plans."
This skill implements the Ralph Loop paradigm - continuous iteration until machine-verifiable completion, not subjective "I think I'm done."
Why Ralph Loop?
Traditional AI assistance fails because:
- Premature exit: AI stops when it feels done, not when it is done
- Context breaks: Restarting loses all progress
- Subjective completion: "Looks good to me" isn't verifiable
Ralph Loop solves this:
- External state as memory: Git commits, test results, file changes persist across iterations
- Machine-verifiable exit: Tests pass? Lint clean? Security audit green? THEN done.
- Continuous iteration: Keep working until objective criteria met
The Core Loop
┌─────────────────────────────────────────────┐
│ │
│ ┌─────────┐ ┌─────────┐ ┌─────────┐ │
│ │ Build │───▶│ Check │───▶│ Pass? │ │
│ └─────────┘ └─────────┘ └────┬────┘ │
│ ▲ │ │
│ │ No │ │
│ └─────────────────────────────┘ │
│ │ Yes │
│ ┌──────▼────┐ │
│ │ DONE │ │
│ └───────────┘ │
└─────────────────────────────────────────────┘
I don't exit because I think I'm done. I exit when:
- ✓ All tests pass
- ✓ Lint returns 0
- ✓ Build succeeds
- ✓ Security audit clean
- ✓ Docs exist
You give me a spec. I build it. I test it. I secure it. I document it. You come back to software that works.
When to Use
- Clear requirements exist (written or verbal)
- You trust me to make reasonable decisions
- You have time away (hours, not minutes)
- You want working software, not status updates
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.
- 9d ago First seen · 408 lines · 30 tokens per session scan A fb291de03097
dev-full-auto is a skill published in the GitHub repository ToruAI/toru-claude-agents (15 stars, last pushed 1mo ago), licensed MIT. It adds 30 tokens to every session and 2,560 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
ln-23-test-suite-auditor
Audits existing tests for meaningful coverage, trustworthy oracles, and maintenance value. Not for test implementation or a single delivery review.
ln-21-documentation-auditor
Audits documentation and comments for trustworthy claims, coverage, and discoverability. Not for code, test, or architecture audits.
ln-41-test-strategy-planner
Plans a risk-based test portfolio and prioritized scenarios without editing tests. Not for test execution or implementation.
ln-42-acceptance-test-builder
Builds, updates, consolidates, or retires scoped acceptance tests and records execution evidence. Not for product-code fixes or audits.
journey-simulation
Use when caller wants to observe how a stranger encounters a flow, artifact, or sandbox — triggers like "simulate a user journey", "test our onboarding / checkout / signup", "will my ICP convert", "how does a cold reader experience this README", "first-time user test", "cognitive walkthrough", or any request to…
test-quality
Write high-quality JUnit 5 tests with AssertJ assertions. Use when user says "add tests", "write tests", "improve test coverage", or when reviewing/creating test classes for Java code.