Agent skills and plugins

GaZmagik/

iso-24495

Seven agent skills that apply ISO 24495 plain language principles to writing, legal, technical, and code output.

What’s new here

This skill pack targets the words an agent produces. It encodes principles inspired by ISO 24495, a published plain language standard series, into seven skill files and a rule engine that measures Markdown against concrete criteria: sentence length, paragraph length, legalese, heading depth, undefined acronyms, and more. The README says the project is unofficial and that every quantitative rule is its own proxy, not a clause of any standard. Parts 4 and 5 of the standard are still unpublished drafts, so those two skills are marked provisional and will be revised when ISO publishes.

What it does

The core skill (iso-24495-1) governs all user-facing output: no filler preambles, short sentences, active voice, scannable structure. It triggers specialist skills automatically: iso-24495-2 for legal text, iso-24495-3 for technical writing, and iso-24495-5 for complex documents.

iso-24495-code applies plain language principles to source code itself: naming, ordering, comments, and error messages. iso-24495-text-audit is a user-invoked skill that runs the rule engine against a .md, .markdown, or .txt file or directory and reports findings with file, line, rule, and explanation. It requires Bun and never modifies files.

The plugin also ships an output style file (output-styles/iso-24495.md) for Claude Code. Select it with /output-style to hold every response to the core rules without relying on skill activation. Codex CLI gets an equivalent skill (iso-24495-style) because Codex has no output style mechanism.

Who it’s for

Technical writers, legal teams, and developers who want an agent to produce consistently plain output. Also useful for documentation maintainers who want automated audits of existing Markdown files. The text audit is English-only; the skills and output style are not limited to English in the same way, since they are instructions a model interprets.

Try it

Install in Claude Code:

/plugin marketplace add https://github.com/GaZmagik/iso-24495.git
/plugin install iso-24495-plain-language@iso-24495

Install in Codex CLI:

codex plugin marketplace add https://github.com/GaZmagik/iso-24495.git
codex plugin add iso-24495-plain-language@iso-24495

To enforce the core skill on every response, add to CLAUDE.md:

- ALWAYS activate and adhere to the `iso-24495-1` Plain Language skill across all responses

Run the text audit on a file (it requires Bun):

/iso-24495-plain-language:iso-24495-text-audit docs/policy.md

Run the test suite locally (requires Bun):

bash scripts/check.sh

How mature is it

191 stars, 8 forks, 1 contributor, 9 releases, latest v0.7.0 on 2026-09-27, created 2026-08-11, last pushed 2026-10-10. At least 100 commits in the last 90 days. 16 open issues and pull requests. MIT license. The project self-audits: the Part 4 maturity checker runs against the repository’s own documents as part of the CI gate.