author

A scaffolding skill for creating AI-layer files such as rules, skills, hooks, agents, and output styles in the repository’s expected format. It verifies the generated file with the relevant format checks.

In plain words
What is it for?
Use it when adding a new .claude artifact and need a conforming template plus checks for its format or shell safety.
Why use it?
It reduces the risk of creating configuration files with the wrong structure or missing required fields.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/ashkaan/contextium/author
Any agent
npx skills add Ashkaan/contextium --skill author
Clone the repo
git clone --depth 1 https://github.com/Ashkaan/contextium

Made for: Claude Code, Codex.

Per session 56 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 4,314 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00056 $0.04314
Opus 5 $0.00028 $0.02157
Sonnet 5 $0.00011 $0.00863
Haiku 4.5 $0.00006 $0.00431

Measured 2d ago against content hash 69bdd639e58c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

author 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.

The scan reads SKILL.md. This mod also ships 12 executable files (references/templates/hook-precommit.template.sh, references/templates/hook-tooluse.template.sh, scripts/grep-rule-peers.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

templates/claude/skills/author/SKILL.md · 292 lines

How it starts

The opening of the file, as written. The whole thing — 292 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/author — Unified AI-Layer Artifact Scaffolder

One skill, five branches. /author <type> [name] scaffolds a conforming .claude/ artifact that already passes the repo's format checks, then runs that check as a backstop. AI judgment is confined to CONTENT (filling a skeleton), never STRUCTURE (the scaffold is deterministic). Per-branch detail lives in references/<type>.md — loaded on demand (progressive disclosure), not inlined.

<type> ∈ {rule, skill, hook, agent, output-style}. The flow is uniform across all five; the type only changes which template, which questions, and which backstop.

Type Branch doc What scaffold writes verify backstop
rule references/rule.md nothing — inline ## <slug> (prints guidance) run-rule-linters.sh
skill references/skill.md .claude/skills/<name>/SKILL.md check-skill-format.sh
hook references/hook.md .claude/hooks/<name>.sh (or checks/) shellcheck + safe-mode
agent references/agent.md .claude/agents/<name>.md frontmatter fields (SSOT in agent.md)
output-style references/output-style.md .claude/output-styles/<name>.md documented-field check (unknown key, description, keep-coding-instructions)

The four principles (enforced, not advised)

Every artifact /author emits must satisfy four principles from the original ask. They are deterministic verify.sh gates (@rule:mechanisms-not-prose), so a violation fails the check — it is not left to judgment. Sourced from Anthropic docs (skills, sub-agents, hooks).

Principle How it's enforced
Anthropic best practices verify.sh: skill description ≤1,536 + third-person; hook exit-2-for-PreToolUse (emitted by placement); agent 6-field SSOT + scoped tools
Low token usage verify.sh: description within the always-loaded cap; SKILL.md body ≤500 lines
High determinism scaffold.sh emits correct-by-construction (right hook shape + exit code per placement); every type has a non-zero-on-violation backstop; skill branch's determinism inventory pushes data steps to scripts/
Low context usage verify.sh: SKILL.md body ≤500 lines; references/ one level deep (leaf docs)

Read the full file on GitHub · 292 lines

Changes

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.

  1. 2d ago First seen · 292 lines · 56 tokens per session scan A 69bdd639e58c

Subscribe to this mod's changes

author is a skill published in the GitHub repository Ashkaan/contextium (32 stars, last pushed 5d ago), licensed MIT. It adds 56 tokens to every session and 4,314 once invoked, about $0.0003 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.