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 agentmods add skills/reflexioai/claude-smart/reviewnpx skills add ReflexioAI/claude-smart --skill reviewgit clone --depth 1 https://github.com/ReflexioAI/claude-smartWhat 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 | $0.00052 | $0.03186 |
| Opus 5 | $0.00026 | $0.01593 |
| Sonnet 5 | $0.00010 | $0.00637 |
| Haiku 4.5 | $0.00005 | $0.00319 |
Grade A, and why
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 yesterday.
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.
Copies of this mod
1 near-identical copy found in the catalogue:
- review — 100% identical, 0 lines differ
How it starts
The opening of the file, as written. The whole thing — 261 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Review
Perform a rigorous, senior-engineer-level code review of all uncommitted changes in the working tree.
Core Principles
- You are a strict reviewer, not a rubber-stamper. Flag real problems. Do not praise code just to be nice.
- Never make changes directly. Your output is a review report and an optional change plan. Wait for explicit user approval before editing any file.
- Embed assumptions inline. If you cannot tell whether something is intentional or a mistake, note your assumption in the report and flag it for confirmation.
- Focus on substance over style. Formatting issues caught by pre-commit hooks are low priority. Lint violations and type errors in changed files are substantive findings — classify them by severity alongside manual review findings.
Review Depth
By default, run the full checklist. If the user requests a quick review (e.g., /review --quick), focus only on:
- Security
- Correctness & Logic
- API & Contract Design
- Error Handling & Resilience
Skip deeper analysis sections (Duplication, Missing Tests, Code Clarity) in quick mode.
Workflow
Phase 1 — Gather the diff
If git diff HEAD produces no output and git status shows no uncommitted changes and no untracked files, inform the user that there are no changes to review and stop.
Run these commands to understand the full scope of uncommitted work:
# Overview of changed files
git status
# Full diff of all tracked changes (staged + unstaged)
git diff HEAD
# List of untracked files that may need review
git ls-files --others --exclude-standard
Read the diff carefully. For every changed file, also read the full file (not just the diff hunk) so you understand the surrounding context — imports, class hierarchy, sibling functions, and call sites.
Run automated checks on changed files (language-aware):
- Detect languages from changed file extensions.
- For Python files (
*.py): runruff check <files>andpyright <files>. - For TypeScript/JavaScript files (
*.ts,*.tsx,*.js,*.jsx,*.mts): runnpx tsc --noEmitandnpx biome check <files>from the relevant project root (reflexio/website/orreflexio/public_docs/). - Skip linters for languages without configured tooling.
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.
- yesterday First seen · 261 lines · 52 tokens per session scan A 09caac7850b4
review is a skill published in the GitHub repository ReflexioAI/claude-smart (774 stars, last pushed 3d ago), licensed Apache-2.0. It adds 52 tokens to every session and 3,186 once invoked, about $0.0003 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.
Other skills, from other repositories
reflexio-embedded
Captures user facts and procedural corrections into .reflexio/ so the agent learns across sessions. Use when: (1) user states a preference, fact, config, or constraint; (2) user corrects the agent and confirms the fix with an explicit 'good'/'perfect' or by moving on without re-correcting for 1-2 turns; (3) at start…
reflexio-consolidate
Run a full-sweep consolidation over all .reflexio/ files — TTL sweep + n-way cluster merge. Use when the user asks to 'clean up reflexio', 'consolidate memory', 'deduplicate playbooks', or suspects drift across sessions.
reflexio
When to use: Always active. This skill is the contract between you and the reflexio cross-session memory plugin.
learn
When to use: The user wants immediate extraction of skills/preferences from the current session — e.g., after a major correction, before context compaction, or to test the loop.
clear-all
When to use: The user explicitly asks to delete all locally-stored skills/preferences. This is destructive and unrecoverable.
show
When to use: The user wants to see what reflexio currently knows about this project.