Borrowing it
Nothing to install: this file belongs to DrDustinEdwards/capsid-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/DrDustinEdwards/capsid-mcp/master/CLAUDE.mdgit clone --depth 1 https://github.com/DrDustinEdwards/capsid-mcpWrote 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/drdustinedwards/capsid-mcp/claude-md)<a href="https://agentmods.dev/instructions/drdustinedwards/capsid-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/drdustinedwards/capsid-mcp/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/drdustinedwards/capsid-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/drdustinedwards/capsid-mcp/claude-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.02409 | $0.02409 |
| Opus 5.5 | $0.00964 | $0.00964 |
| Sonnet 5.5 | $0.00482 | $0.00482 |
| Haiku 4.5 | $0.00241 | $0.00241 |
Grade A, and why
capsid-mcp 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 1mo 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md - capsid-mcp
This file is the briefing you need BEFORE you can reach Capsid. Everything else is in Capsid. Consolidated 2026-08-14 from 13.7KB: the precedence model (capsid/repo-structure.md, layer 8) bans a repo file from restating canon, and this one had grown into a second canon store, holding its own copies of the auth model, the backup runbook and a defect list that Capsid already carried and kept more current.
What this is
Capsid: a single-user, Cloudflare-native MCP server serving a consolidated knowledge base from D1 and R2, reaching the GitHub repos directly. The memory and CMS layer for the whole portfolio. Live at https://capsid.dustin-edwards.workers.dev/mcp. Single-user per deployment by design; multi-tenant is out of scope.
Capsid documents itself, and those documents outrank this file. Read them before trusting any description of this system, including this one:
capsid/conventions.mdportfolio-wide standing rules. Read first, every session.capsid/core.mdwhat Capsid IS and its current status. The only place status lives: tool count, deployed sha, open items.capsid/decisions.mdthe rulings and why they are what they are.capsid/schema.mdthe knowledge model and the lint loop.capsid/repo-structure.mdthe 9-layer precedence model and the .claude standard shape.capsid/protocol-restore.mdtested restore procedures, including the D1 export gotcha.
Session ritual
Start: read capsid/conventions.md, then capsid/core.md.
End: write NOTHING by default. The end-of-session episodic was withdrawn portfolio-wide 2026-08-21 (capsid/conventions.md); a session records a REVERSAL or a BINDING to capsid/decisions.md, or nothing.
Reaching it
- MCP tool lists cache at connect time. A tool deployed mid-session is invisible until the connector reconnects or a new chat starts. Verified again 2026-08-14, when a reconnected session still held 22 tools against a deployed 24.
- To verify a fresh deploy without waiting for a reconnect, call the Worker directly: POST
initialize, thennotifications/initialized, thentools/call, parsing SSEdata:lines. The OAuth access token is in~/.claude/.credentials.jsonundermcpOAuth(key prefixcapsid|). Never print it. - Two gated paths: OAuth on
/mcpfor human clients, agent and operator keys on/ops/mcpfor agents and cron. Every caller resolves to a scoped agent and every tool call goes through one enforcement point; the legacy operator key keeps its authority until its hash is removed by hand. The model is in README.md under Auth model; minting one isdocs/bootstrap.md.
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.
- 1mo ago Changed · +2 lines · +513 tokens per session 7053e0272b42
- 1mo ago Changed · +1 lines · +259 tokens per session 553637f9c3f8
- 1mo ago Changed · +1 lines · +123 tokens per session 3aea003b24ab
- 1mo ago Changed · +2 lines · +269 tokens per session df1a829a249c
- 1mo ago First seen · 56 lines · 1,245 tokens per session scan A ff79902b7aac
capsid-mcp CLAUDE.md is an instructions file published in the GitHub repository DrDustinEdwards/capsid-mcp (0 stars, last pushed 1mo ago), licensed MIT. It adds 2,409 tokens to every session, about $0.0096 per session on Opus 5.5. 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
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.