Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Internal Shipped Reference

ASecurity

Reviews a shipped-corpus edit for a reference that resolves only in this checkout and reads as portable, the one form the reference gate cannot classify by pattern. Use before shipping a change under claude/, docs/, governance/, scripts/, snippets/, standards/, or tooling/, routed here by `598-authoring-layout.md`. Do NOT use for a pull request number, a commit sha, a resolvable `docs/<name>.md` path, a phase label, or a numbered rule path, which `canon gate run`'s shipped-references stage al...

2 stars
0 votes
0 copies
0 views
Added 9/20/2026
developmentgo

Works with

cli

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add erclx/canon --skill internal-shipped-reference --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Internal Shipped Reference?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Internal Shipped Reference
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/erclx-internal-shipped-reference/badge)](https://www.skillsdirectory.com/skills/erclx-internal-shipped-reference)

More formats (shields.io, HTML) on the badges page.

Download with Pro
Files
SKILL.md
---
name: internal-shipped-reference
description: Reviews a shipped-corpus edit for a reference that resolves only in this checkout and reads as portable, the one form the reference gate cannot classify by pattern. Use before shipping a change under claude/, docs/, governance/, scripts/, snippets/, standards/, or tooling/, routed here by `598-authoring-layout.md`. Do NOT use for a pull request number, a commit sha, a resolvable `docs/<name>.md` path, a phase label, or a numbered rule path, which `canon gate run`'s shipped-references stage already classifies by pattern.
---

# Internal shipped reference

`src/shipped/references.ts` classifies five same-repository reference forms by pattern and stops there on purpose. Its own `DOCS_PATH` comment names the sixth: telling a `canon/context/<name>.md` citation of this repository's own domain entry apart from an illustration of a target's own generic tree, such as `canon/context/index.md`, is a semantic read no pattern makes. This skill is that read, plus the parallel judgment over a layout or count claim.

## Discriminator

Read every added or changed line in the diff under this rule's eight corpora: `internal/`, `standards/`, `snippets/`, `claude/`, `governance/`, `docs/`, `scripts/`, `tooling/`.

For each `canon/context/<name>.md` token:

- **Fine.** The sentence reads as an instruction for a target session to consult its own project's copy if one exists, or illustrates the path shape a target's own tree would take, never naming a fact only this checkout holds. `ux-measure/SKILL.md`'s `canon/context/development/: the documented run commands and the port each serves` is this case, read as "check your own project's entry."
- **Fails.** The sentence names a fact true of this repository specifically, one only this checkout's own `canon/context/` resolves, such as a documented decision or a domain narrative a target reader cannot open. `canon-cli/SKILL.md`'s prior citation of `canon/context/tooling.md` as "the toolkit's" was this case.

Do not test a token against what a scaffolded project's seed happens to ship. That discriminator was measured and rejected: the seed's own contents drift, and a token reading as an instruction is fine regardless of what any seed carries.

A token inside a branch only a toolkit-repo reader ever follows, such as an `In the toolkit:` clause beside an `In a project:` one, is Fine even when it names a domain entry, since the reader who reaches that branch is by construction sitting in this checkout. `create-standard/SKILL.md`'s `In the toolkit:` line is this case. An unbranched sentence naming the same path for every reader stays a Fails, since a target reader reaches it too.

For each layout, stack, or config-path claim:

- **Fine.** The claim names a portable convention true of any project carrying the same toolkit surface, such as a fixed install path (`canon/config/pr-labels.toml`).
- **Fails.** The claim states a count, a stack choice, or a folder shape true of this checkout specifically, phrased as if every target shared it.

## Repair menu

- State the fact the reference carries instead of the path. `claude/skills/canon-cli/SKILL.md`'s repaired form, naming the toolkit's own context entries as authoritative over a target-session summary without the three paths, is the shape.
- Where the sentence labels the path toolkit-only inside text every reader sees, the label is not the fix. Drop the path and fold what it pointed at into the sentence, or drop the trailing clause when the path carried nothing past the label. `youtube-transcripts/SKILL.md`'s repaired form is this case.
- Where the path sits inside a branch only a toolkit-repo reader ever follows, keep it and mark `<!-- canon-allow-reference: <why> -->`, naming the branch. That reader can open the path, so dropping it costs the only audience the sentence has. `create-standard/SKILL.md`'s `In the toolkit:` branch is this case, not the bullet above.
- Where the token is a genuine illustration rather than a citation, mark it the same way, the marker `src/shipped/references.ts` already reads for its own four forms. Nothing in the gate scans a `canon/context/` token today, so the marker documents the read for the next reader rather than suppressing a check.
- Scope a checkout-specific claim to "in the toolkit" or "in this repository" rather than deleting it, when the fact is worth keeping for a toolkit-only reader.

## Report

State each finding as `<file>:<line>: <token>`, which half of the discriminator it failed, and the repair applied or proposed. Report a clean pass as `No self-only reference found` rather than staying silent, so a reviewer reading the diff sees the skill ran.

Attribution

erclxerclx
View sourceMore from erclx →
SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Know which skills are safe — weekly.

Best new skills + every skill we flagged as malicious. From the team that scanned 103,619.

Join free

Related Skills

Browser Extension Developer

Use this skill when developing or maintaining browser extension code in the `browser/` directory, including Chrome/Firefox/Edge compatibility, content scripts, background scripts, or i18n updates.

284722 votes

Seo Optimizer

SEO optimization with keyword analysis, readability assessment, technical validation, content quality. Use for search rankings, blog posts, content audits, or encountering keyword density, readability scores, meta tags, schema markup errors.

2192 votes

Google Official Seo Guide

Official Google SEO guide covering search optimization, best practices, Search Console, crawling, indexing, and improving website search visibility based on official Google documentation

1862 votes

Tanstack Start

Build a full-stack TanStack Start app on Cloudflare Workers from scratch — SSR, file-based routing, server functions, D1+Drizzle, better-auth, Tailwind v4+shadcn/ui. Use whenever the user mentions TanStack Start, asks to scaffold a full-stack Cloudflare app with SSR, wants an SSR dashboard, or asks for a React 19 + Cloudflare Workers app with file-based routing and server functions — even if they don't name TanStack Start specifically. No template repo — Claude generates every file fresh per ...

9881 votes

Pentest

PTES-aligned adversarial security audit for backend, frontend, and mobile applications. Produces a CVSS-scored Hacker Report with verified PoCs and phased remediation.

5491 votes
View all in development →