Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/kangig94/coralnpx agentmods add skills/kangig94/coral/reefWrote 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/kangig94/coral/reef)<a href="https://agentmods.dev/skills/kangig94/coral/reef"><img src="https://agentmods.dev/badge/skills/kangig94/coral/reef.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Rogue Agent · line 45 Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.Fix: Remove any persistence mechanisms (cron jobs, startup scripts, state files). Skills should not maintain state across sessions without explicit user consent.
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 | $0.00017 | $0.01165 |
| Opus 5 | $0.00009 | $0.00583 |
| Sonnet 5 | $0.00003 | $0.00233 |
| Haiku 4.5 | $0.00002 | $0.00117 |
Grade A, and why
reef 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 7d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s http://localhost:3100/api/system/health How it starts
The opening of the file, as written. The whole thing — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Coral Reef Dashboard
ARCHIVED — not a live skill. Moved out of
skills/so it is not discovered or exposed to users (Claude Code recursively registers everyskills/<name>/SKILL.md; there is no frontmatter/naming flag to hide one in place).coral-reefdevelopment is paused. To re-enable: move this directory back toskills/reef/. Note the data dir (~/.claude/coral-reef/) still hardcodes~/.claudeand does not yet honorCLAUDE_CONFIG_DIR— fix that in thecoral-reefserver repo before re-enabling.
Launch the coral-reef dashboard. Handles installation, updates, builds, and startup automatically.
The runtime injects Base directory for this skill: which resolves to this skill's installed path.
From there, the plugin root is two levels up, and coral-reef/ sits as a sibling directory.
Constants
REPO_URL = "https://github.com/kangig94/coral-reef"
REEF_DATA_DIR = ~/.claude/coral-reef/
REEF_PORT = 3100
Protocol
Step 1 — Resolve Paths
Derive the plugin root from the base directory injected by the framework:
PLUGIN_ROOT = resolve(base_directory, "../..")
REEF_DIR = resolve(PLUGIN_ROOT, "coral-reef")
Verify PLUGIN_ROOT contains package.json with "name": "coral" as a sanity check.
Step 2 — Ensure Data Directory
mkdir -p ~/.claude/coral-reef
This is coral-reef's permanent SQLite storage location, surviving /tmp reboots.
Step 3 — Clone or Update
Check if REEF_DIR exists:
- Not exists:
git clone ${REPO_URL} ${REEF_DIR} - Exists with
.git/:cd ${REEF_DIR} && git pull --ff-only - Exists without
.git/: warn user and skip update (manual install)
Step 4 — Dependency Resolution
Detect the install layout by checking if this is a nested plugin install or a dev workspace:
# Read coral-reef's package.json
cd ${REEF_DIR}
Check if the coral dependency in package.json points to the correct location:
- Dev workspace layout (
~/workspace/coral-reefsibling to~/workspace/coral): Thecoraldependency should be"file:../coral". Leave as-is.
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.
- 7d ago First seen · 136 lines · 17 tokens per session scan A 1515432ab7b6
reef is a skill published in the GitHub repository kangig94/coral (11 stars, last pushed 5d ago), licensed MIT. It adds 17 tokens to every session and 1,165 once invoked, about $0.0001 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 2026-08-30.
Other skills, from other repositories
accessibility
This skill should be used when building or reviewing UI for accessibility — keyboard navigation, screen readers, ARIA, color contrast, focus states, accessible forms, or WCAG compliance. Trigger phrases include "make it accessible", "a11y", "accessibility", "WCAG", "ARIA", "keyboard navigation", "screen reader"…
frontend-performance
This skill should be used when a frontend feels slow, fails Core Web Vitals, has a large bundle, or needs performance optimization. Trigger phrases include "make it faster", "improve performance", "Core Web Vitals", "LCP", "INP", "CLS", "reduce bundle size", "code splitting", "lazy load", "it's slow to load"…
landing-page
This skill should be used when the user wants to create a landing page, sales page, home page, lead-capture page, or any conversion-oriented page. Trigger phrases include "build a landing page", "make a sales page", "create a home page for my product", "I need a page to sell X", "landing for my SaaS/course/service"…
pwa
This skill should be used when turning a web app into a Progressive Web App — making it installable, work offline, cache assets, or send push notifications. Trigger phrases include "make it a PWA", "installable app", "work offline", "offline mode", "service worker", "web push", "push notifications", "add to home…
design-taste-frontend
Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.
image-to-code
Elite website image-to-code skill for Codex. For visually important web tasks, it must first generate the design image(s) itself, deeply analyze them, then implement the website to match them as closely as possible. In Codex, it must prefer large, readable, section-specific images instead of tiny compressed boards…