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/a8cteam51/claude-code-plugins/annotatenpx skills add a8cteam51/claude-code-plugins --skill annotategit clone --depth 1 https://github.com/a8cteam51/claude-code-pluginsWhat 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 | $0.00160 | $0.05721 |
| Opus 5 | $0.00080 | $0.02861 |
| Sonnet 5 | $0.00032 | $0.01144 |
| Haiku 4.5 | $0.00016 | $0.00572 |
Grade A, and why
annotate scanned grade A with 1 finding 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
Playwright, curl, or any other mechanism. How it starts
The opening of the file, as written. The whole thing — 474 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Annotate a live web page into GitHub issues
Arm an annotation overlay in the page the user is viewing in Chrome, wait while they click elements and attach notes, then turn each annotation into a GitHub issue with a screenshot. This is a front-end QA companion: instead of describing a visual bug in words, the user points at the element in the real, logged-in, fully-rendered page, and the finding lands in a tracker.
One annotation → one issue. This skill does not read or edit the codebase.
The overlay lives in the browser as a Tampermonkey userscript, not as an
injected payload. This plugin owns the canonical copy at
assets/page-annotator.user.js (relative to this skill's directory) and is
responsible for installing it and keeping the installed version in sync.
Two facts drive everything below:
- The userscript runs in Tampermonkey's sandbox;
javascript_toolruns in the page's main world. They share only the DOM. So every exchange is a DOM read or an attribute write — never awindowglobal. - The userscript is idle on every page until armed. It stamps its version
on
<html>at document-start and waits.
| Marker | Direction | Meaning |
|---|---|---|
<html data-claude-annotator="0.3.0"> |
script → you | Userscript installed, at that version |
<html data-claude-annotator-ack="…"> |
script → you | Echo of the last command handled |
<html data-claude-annotate="…"> |
you → script | on | off | shot:<id> | shot:end |
<html data-claude-annotate-config="…"> |
you → script | JSON: repo prefill and filed issue numbers |
<div id="__claude_annotator_host__"> |
— | The overlay itself — present iff it is armed |
<script id="__claude_annotations__"> |
— | The payload; outlives the overlay on cancel |
That last distinction matters: clicking ✕ removes the host but leaves the
state node behind holding status: "cancelled", so the node is not a reliable
"is it armed" signal. Always test the host for that.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 474 lines · 160 tokens per session scan A 8234fc889ce9
annotate is a skill published in the GitHub repository a8cteam51/claude-code-plugins (3 stars, last pushed 4d ago), licensed MIT. It adds 160 tokens to every session and 5,721 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
babysit-pr
Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…
imagegen
Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…
cpu-profile-analysis
Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…