Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add MasonEgger/bpe-claude-code-plugin/plugin install bpeWrote 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/masonegger/bpe-claude-code-plugin/review)<a href="https://agentmods.dev/skills/masonegger/bpe-claude-code-plugin/review"><img src="https://agentmods.dev/badge/skills/masonegger/bpe-claude-code-plugin/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/skills/masonegger/bpe-claude-code-plugin/review"><img src="https://agentmods.dev/badge/skills/masonegger/bpe-claude-code-plugin/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 | $0.00044 | $0.03825 |
| Opus 5.5 | $0.00018 | $0.01530 |
| Sonnet 5.5 | $0.00009 | $0.00765 |
| Haiku 4.5 | $0.00004 | $0.00382 |
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 1mo 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 — 210 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Review Command
Generate a styled, interactive HTML view of spec.md, plan.md, or todo.md and serve it on a local HTTP server for human review. The artifact is broken into fine-grained decision units; the user marks each one Ship it / Close, but update / Completely off, do this instead / Absolutely reject and delete, leaves comments, and clicks Save. Feedback is written to JSON for /bpe:apply-review to consume.
The server runs in the background; this command does not block the Claude Code terminal. Continue working or wait; either is fine.
Step 1: Determine the Artifact
Resolve the target artifact:
- If the user passed an argument, map it:
spec→spec.md,plan→plan.md,todo→todo.md. - If no argument was passed, pick the most recently modified of
spec.md,plan.md,todo.mdin the project root. - If none exist, tell the user and stop.
Step 2: Generate the HTML Artifact
Read the chosen markdown file and render it into the fixed template below. Your job is to fill the template with content, not to invent layout or typography; the design is owned by review.css, which the server inlines into the page. Emitting the semantic classes from the contract is what keeps every review page consistently readable. Do not restyle the chrome (.review-toc, .review-section, .decision, .review-controls) with your own CSS or Tailwind utilities; that reintroduces the per-run drift the template exists to prevent.
Decompose into the smallest decidable units
A decision unit is the smallest thing the reviewer would want to accept, tweak, redirect, or kill on its own. One decision unit = one <section> with one set of decision buttons. Do not bundle. If an artifact heading like "Mode-by-mode changes" describes three modes, that is three decision units (one per mode), not one; bundling them forces a single verdict on independent ideas and is exactly what makes review painful.
Err aggressively toward more units. The test: "Could the reviewer plausibly ship one part and say 'completely off, do this instead' to the next?" If yes, split them. Every distinct proposal, design choice, open question, or assumption you'd want a ruling on becomes its own unit with its own buttons. Many small units is the goal, not a side effect.
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.
- 1mo ago Changed 8a4cdddd3ec2
- 2mo ago First seen · 210 lines · 44 tokens per session scan A 64ddb6b261e2
review is a skill published in the GitHub repository MasonEgger/bpe-claude-code-plugin (7 stars, last pushed 27d ago), licensed MIT. It adds 44 tokens to every session and 3,825 once invoked, about $0.0002 per session on Opus 5.5. 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
interactive-login
Log a CLI in with a browser/device-code flow (aws sso, gh, glab, gcloud, …) and save the result to the keychain so later commands can request it through execute.credentials.
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
azure-messaging-webpubsub-java
Build real-time web applications with Azure Web PubSub SDK for Java. Use when implementing WebSocket-based messaging, live updates, chat applications, or server-to-client push notifications.
cua-driver
Use Cua Driver for desktop or browser tasks that are awkward or unavailable through Bash/APIs, or when the user explicitly wants GUI interaction: app testing, visual bug reproduction, form filling, calendar entry, screenshots, and demo recording. Also covers Cua setup; not OpenAI Codex Computer Use or web research.
open-design-browser
Inspect and exercise local HTML design artifacts and library prototype previews with the existing agent-browser CLI, including screenshots, interaction checks, and rendered DOM snapshots. Use when rendered evidence helps author or review a design; excludes browser installation, remote services, and unrelated browser…
browser_cdp
A health check for Python code that produces a score, letter grade, and measures of issues such as complexity, duplication, dead code, dependencies, and architecture.