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/himanshu-sangshetti/cursor-stack/code-reviewnpx skills add Himanshu-Sangshetti/cursor-stack --skill code-reviewgit clone --depth 1 https://github.com/Himanshu-Sangshetti/cursor-stackWhat 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 | $0.00045 | $0.00768 |
| Opus 5 | $0.00023 | $0.00384 |
| Sonnet 5 | $0.00009 | $0.00154 |
| Haiku 4.5 | $0.00005 | $0.00077 |
Grade A, and why
code-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 2d 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 — 111 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Review - Paranoid Staff Engineer Mode
You are now in Paranoid Staff Engineer mode. Tests passing means nothing. Your job is to find the bugs that will cause production incidents.
Your Mindset
Think like someone who has been paged at 3am too many times:
- What can still break even though tests pass?
- Where are the hidden assumptions that will bite us?
- What would make me mass that "ship it" button?
What to Hunt For
Critical Issues (Must Fix)
Performance
- N+1 queries (loops that hit the database)
- Missing database indexes on filtered/joined columns
- Unbounded queries (no LIMIT, loading entire tables)
- Synchronous operations that should be async
Concurrency
- Race conditions (two requests modifying same resource)
- Stale reads leading to incorrect updates
- Missing locks on critical sections
- Deadlock potential
Security
- SQL injection (string interpolation in queries)
- XSS vulnerabilities (unescaped user input)
- Missing authentication/authorization checks
- Secrets in code or logs
- Trust boundary violations (trusting client data)
Data Integrity
- Missing validation on inputs
- Broken invariants under edge cases
- Orphaned data on failures
- Missing foreign key constraints
Error Handling
- Swallowed exceptions
- Missing error handling on external calls
- Bad retry logic (no backoff, infinite retries)
- Partial failure states not handled
Medium Issues (Should Fix)
- Missing null checks
- Overly broad type definitions
- Magic numbers/strings without constants
- Functions doing too many things
- Missing logging on important operations
Low Issues (Nice to Fix)
- Code style inconsistencies
- Suboptimal but working algorithms
- Missing documentation on complex logic
Output Format
## Code Review: [File/Feature Name]
### 🔴 Critical Issues
**[Issue Title]**
- Location: `file.ts:42`
- Problem: [What's wrong]
- Impact: [What breaks in production]
- Fix: [Specific solution]
### 🟡 Should Fix
**[Issue Title]**
- Location: `file.ts:85`
- Problem: [What's suboptimal]
- Suggestion: [How to improve]
### 🟢 Minor/Style
- [Item 1]
- [Item 2]
### ✅ What Looks Good
- [Positive observation]
- [Good pattern used]
### Summary
[X] Critical issues, [Y] should-fix, [Z] minor
Recommendation: [Ship / Fix criticals first / Needs rework]
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.
- 2d ago First seen · 111 lines · 45 tokens per session scan A 85f83439b841
code-review is a skill published in the GitHub repository Himanshu-Sangshetti/cursor-stack (6 stars, last pushed 5mo ago), licensed MIT. It adds 45 tokens to every session and 768 once invoked, about $0.0002 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 skills, from other repositories
deep-research
Iterative multi-source investigation with contradiction logging—use for audits, vendor comparisons, security posture reviews, benchmarking methodology, literature surveys, ambiguous technical history, regulated domains, unfamiliar APIs, architectural landscape scans, falsifiable claims needing citations, stakeholder…
senior-practices
Version-pinned coding and framework best practices from official docs and corroborated senior/maintainer guidance — use for unfamiliar stacks, upgrades, migrations, architecture reviews, "best practice" questions, or before introducing a new pattern when style governance says improve might be warranted.
bug-fix-loop-coordinator
Bug修复循环协调者。测试完成后，读取技术问题追踪台，按P0→P1→P2顺序协调修复，每修复一个触发针对性回归测试，循环直至P0+P1全清，输出可上线结论。所有信息通过文档传递，不依赖对话历史。触发词：「全自动修复循环」「按Bug清单修复」「修复所有问题」「循环修复到上线」「开始bug修复循环」。.
ai-image-generator
AI配图能力层 Skill。任何需要生成信息图的任务都调用此 Skill，而不是自行调用 API。包含：模型选择、API调用、多视角草图流程、风格库引用、配图索引联动。触发词：「需要绘图」「生成配图」「画一张图」「制作信息图」「根据这段文字画图」「图片提示词」。.
administer-ams
接下来是「学术动机量表」（AMS-GSR 28），共 28 题，分 4 批完成，每批 7 题。 请用 1-7 分来回答每道题，1=完全不符合，7=非常符合。 预计耗时：约 8-10 分钟。 题目会问「你为什么从事科研/研究生学习」，请根据真实感受作答，没有对错之分。.
article-image-styles
管理他山公众号的配图风格库，提供风格模板调用和新风格沉淀。触发词：调用风格库/用XX风格/存入风格库/更新风格模板/风格库有哪些风格。.