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/guyoung/boxagnts/code-reviewnpx skills add guyoung/boxagnts --skill code-reviewgit clone --depth 1 https://github.com/guyoung/boxagntsWhat 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 | $0.00024 | $0.02134 |
| Opus 5 | $0.00012 | $0.01067 |
| Sonnet 5 | $0.00005 | $0.00427 |
| Haiku 4.5 | $0.00002 | $0.00213 |
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 — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GUIDANCE SKILL — INSTRUCTIONS ONLY Code review is a systematic examination of source code quality across multiple dimensions: logic, security, performance, maintainability, and standards compliance. This skill provides patterns, conventions, and constraints. It contains no executable code — apply the guidance to your context, adapting specifics as needed per C2-B.
CRITICAL: Guidance Principles
The instructions below define the REQUIRED APPROACH — you MUST follow the prescribed patterns, conventions, and constraints.
Mandatory adherence:
- Follow all stated conventions and naming patterns
- Respect all constraints and boundaries
- Apply the recommended patterns to your implementation
- Do NOT deviate from prescribed architectural decisions
- Never modify any code; the review is strictly read-only
- Every issue reported MUST include a concrete, actionable fix suggestion
Correct Application Pattern
- Read and absorb all guidance sections below
- Map the prescribed patterns to your current context
- Implement following the stated conventions — adapt specifics to fit, but preserve the intent
- Validate against the C3 checklist at the end
Argument Details
| Arg | Type | Required | Default | Description |
|---|---|---|---|---|
| target | string | No | (empty) | File or directory to review; leave empty to review git staged changes |
Design Patterns & Conventions
Architectural Pattern
Multi-dimensional assessment: every review MUST examine the target code through all five review dimensions — Logic Correctness, Security, Performance, Maintainability, and Standards Compliance. Issues from each dimension are independently classified and then merged into a single ranked report.
Naming Conventions
Issues are classified by severity using four standard levels:
| Severity | Meaning |
|---|---|
| Critical | Security vulnerability, data loss risk, or production-breaking bug |
| High | Logic error with significant impact, or major performance regression |
| Medium | Maintainability concern, minor duplication, or unclear naming |
| Low | Style nit, non-critical lint warning, or cosmetic improvement |
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 · 185 lines · 24 tokens per session scan A 50bc97aff1ef
code-review is a skill published in the GitHub repository guyoung/boxagnts (11 stars, last pushed 1mo ago), licensed MIT. It adds 24 tokens to every session and 2,134 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.
Other skills, from other repositories
feature
Use this whenever the user asks to "implement", "add", "build", "create", or "modify" a feature, page, view, component, or module in this Vue project. Three phases: Phase 0 scope analysis (flows + edge cases + UI needs + plan validation, STOP for user), Phase 1 implementation (layered UI → Store → API, Vue 3…
pull-request
Use this whenever the user asks to "open a PR", "ship this", "create the pull request", "make a PR", or once a Vue feature/fix is verified and ready to push. Full lifecycle: branch → commit (commitizen) → issue → draft PR → CI → ready → autonomous monitor loop (fix comments, resolve threads, iterate until CI green +…
naming
Use this whenever a new file/folder is created or renamed in this Vue project, when reviewing a module for consistency, or when the right path for a file is unclear. Also triggers on "what should I name this?", "is this file named correctly?", "audit module naming". Enforces kebab-case folders, dot-prefixed…
ui
Design system reference for the Devkit Vue stack (Vuetify 4 + custom theme helpers). Use when asked about styling, theming, layout, design system, Vuetify components, visual verification, or UI patterns. Automatically consumed by /feature when the task has a visual component. Can also be invoked directly for…
create-module
Use this whenever the user asks to "create a module", "scaffold a feature", "add a Vue domain", "new module called X", or starts work on a brand-new vertical (views + components + store + router entry + tests). Duplicates the canonical src/modules/tasks template, applies kebab/Pascal/camel/ UPPER renames, and wires…
verify
Run this after every code change in this Vue project, before committing, before opening a PR, or whenever the user says "verify", "check", "lint", "run tests", "build", "audit". Diff audit (router guards, store boundaries, UX, /ui consistency, error handling) → lint → tests + coverage gate → build. Coverage drops →…