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 instructions/bguthrie/prospero/claude-mdgit clone --depth 1 https://github.com/bguthrie/prosperoWrote 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/instructions/bguthrie/prospero/claude-md)<a href="https://agentmods.dev/instructions/bguthrie/prospero/claude-md"><img src="https://agentmods.dev/badge/instructions/bguthrie/prospero/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.01565 | $0.01565 |
| Opus 5 | $0.00783 | $0.00783 |
| Sonnet 5 | $0.00313 | $0.00313 |
| Haiku 4.5 | $0.00156 | $0.00156 |
Grade A, and why
prospero CLAUDE.md 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 4d 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.
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.
How it starts
The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
What this repository is
Prospero is a Claude Code plugin (not an application). There is no build step, runtime, or package manager. The "code" is markdown and TOML: command shims, skill prompts, CMS presets, and rubric templates. Claude Code loads everything at runtime from the plugin root.
Plugin manifest: .claude-plugin/plugin.json. Marketplace entry: .claude-plugin/marketplace.json.
Commands
The only automated check is a static lint:
./tests/validate.sh
It verifies the manifest parses, the required skills/commands/presets/templates exist, each preset has its required keys (posts_dir, post_path_pattern, drafts_dir, frontmatter_template), and every SKILL.md has name: and description: frontmatter. Run it after any structural change.
There is no test runner for skill behavior — skills are LLM prompts, not code. tests/fixtures/hugo/{before,after}/ shows the expected filesystem state around /init; examples/sample-post/ shows per-phase artifacts for a single post.
Architecture
The four-phase pipeline
interrogate → critique → author → revise. Each phase is one skill that does one job and produces one artifact; phases communicate via files on disk, not conversation state. The prospero skill is a router that picks a phase from user intent; it does not chain phases or duplicate their preflight checks. Each phase decides its own handoff (e.g., /interrogate auto-invokes /critique; /critique does NOT auto-invoke /author — the author must accept the critique first).
Cross-phase state is files, keyed by slug:
- Outline:
<drafts_dir>/<slug>/outline.md— written by interrogate, read by critique/author. - Research:
<drafts_dir>/<slug>/research.md— the accumulating research record for the post. Interrogate initializes it, critique extends it (and consumes it before doing new web calls — research is not re-derived), author expands it while grounding the draft. Every writer appends under aYYYY-MM-DDTHH:MMsection header (local time, to the minute) so multiple same-day passes do not collide. Never overwritten. - Post: resolved
post_path_pattern— written by author, edited by revise.
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.
- 4d ago First seen · 91 lines · 1,565 tokens per session scan A 797d78ee70e6
prospero CLAUDE.md is an instructions file published in the GitHub repository bguthrie/prospero (6 stars, last pushed 3mo ago), licensed MIT. It adds 1,565 tokens to every session, about $0.0078 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 2026-08-31.
Other instructions, from other repositories
avoid-ai-writing CLAUDE.md
Instructions for conorbronsdon/avoid-ai-writing, covering claude.md, what this is, repository structure, how to make changes and architecture of the skill.
she-love-me CLAUDE.md
Instructions for 863401402/she-love-me, covering claude.md and 唯一工作流.
rolecraft AGENTS.md
Instructions for rolecraft-sh/rolecraft, covering rolecraft — development guidelines, ⚠️ startup checklist (read before every task), workflow, interaction rules and code style.
claude-omakase CLAUDE.md
Claude Code instructions for gyuch-an02/claude-omakase, covering claude.md, commands, architecture (the big picture), conventions and out of scope.
pencil-skill AGENTS.md
Instructions for Nisus74/pencil-skill, covering pencil-dev-skill: project context, project purpose, naming conventions, repository structure and project context files.
skill.color-expert CLAUDE.md
Claude Code instructions for meodai/skill.color-expert, covering claude.md, project overview, architecture, no build/test/lint and editing guidelines.