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.
git clone --depth 1 https://github.com/Smirnov-Labs/claude-skillsWrote 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/commands/smirnov-labs/claude-skills/loop-review)<a href="https://agentmods.dev/commands/smirnov-labs/claude-skills/loop-review"><img src="https://agentmods.dev/badge/commands/smirnov-labs/claude-skills/loop-review/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/commands/smirnov-labs/claude-skills/loop-review"><img src="https://agentmods.dev/badge/commands/smirnov-labs/claude-skills/loop-review.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 | $0.00011 | $0.00971 |
| Opus 5 | $0.00005 | $0.00485 |
| Sonnet 5 | $0.00002 | $0.00194 |
| Haiku 4.5 | $0.00001 | $0.00097 |
Grade A, and why
loop-review 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 11d 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.
How it starts
The opening of the file, as written. The whole thing — 125 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Loop Review Command
You are initiating an iterative plan refinement loop. A senior software architect will review the current plan through multiple rounds until it meets a quality threshold.
Prerequisites
There MUST be a plan in the current conversation context. If no plan exists, tell the user to run /plan first and stop.
Process
Step 1: Capture the Current Plan
Identify the most recent plan from the conversation. This is the plan that was presented to the user — including phases, steps, files, testing strategy, etc.
Step 2: Run the Review Loop
Execute up to 3 review iterations. On each iteration, use the Agent tool with the following:
Agent role: You are a senior software architect (principal-level) performing a rigorous review of an implementation plan. You are critical, precise, and care deeply about quality.
Agent prompt (include the full plan text in each call):
You are a senior software architect reviewing an implementation plan. Be rigorous and critical.
## The Plan to Review
[INSERT FULL PLAN HERE]
## Project Context
Read these files for project context before reviewing:
- CLAUDE.md
- .claude/memory-bank/systemPatterns.md
- .claude/memory-bank/techContext.md
- .claude/memory-bank/activeContext.md
## Review Dimensions
Evaluate the plan on ALL of these dimensions:
1. **Completeness** — Are there missing steps, unhandled edge cases, or gaps?
2. **Sequencing** — Are steps in the right dependency order? Are there parallelism opportunities being missed?
3. **Risk Assessment** — What could go wrong? Are failure modes addressed?
4. **Scope Discipline** — Is there over-engineering, unnecessary complexity, or scope creep?
5. **Pattern Alignment** — Does the plan follow the project's established patterns and conventions?
6. **Testability** — Is the testing strategy adequate? Are success criteria verifiable?
## Output Format
You MUST respond in EXACTLY this format:
### Score: [1-10]
### Must-Fix (blocking issues that must be resolved)
- [issue]: [specific suggestion to fix it]
### Suggestions (non-blocking improvements)
- [suggestion]: [rationale]
### Approved (things done well)
- [good aspect]: [why it works]
### Refined Plan
[If score < 8, provide the COMPLETE refined plan with all fixes applied. Do not provide a partial plan or just the changes — output the full plan ready to use.]
[If score >= 8, write "Plan approved as-is." and do not rewrite it.]
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.
- 11d ago First seen · 125 lines · 11 tokens per session scan A f5b4aa2124ec
loop-review is a command published in the GitHub repository Smirnov-Labs/claude-skills (5 stars, last pushed 2mo ago), licensed MIT. It adds 11 tokens to every session and 971 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 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.