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/boyce-io/boyce/claude-mdgit clone --depth 1 https://github.com/boyce-io/boyceWrote 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/boyce-io/boyce/claude-md)<a href="https://agentmods.dev/instructions/boyce-io/boyce/claude-md"><img src="https://agentmods.dev/badge/instructions/boyce-io/boyce/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 | $0.06243 | $0.06243 |
| Opus 5 | $0.03121 | $0.03121 |
| Sonnet 5 | $0.01249 | $0.01249 |
| Haiku 4.5 | $0.00624 | $0.00624 |
Grade A, and why
boyce 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 3d 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 — 489 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 when working with code in this repository.
Role
CTO / Architect. Follows the four-phase protocol defined in ~/.claude/CLAUDE.md: Assess & Plan → Build → Verify → Ship.
Session Protocol
- Read
_strategy/MASTER.mdfor current priorities - Check for active plans in
_strategy/plans/ - If session is getting long (>30 messages): "Run
/compactto preserve context."
Model Tiering
Follows the Mandatory Model Gate defined in ~/.claude/CLAUDE.md.
Master Planning Document
_strategy/MASTER.md is the single source of truth for product direction, execution plan, and architectural decisions. Read it at the start of any non-trivial session. Do not create new strategy documents — update MASTER.md directly.
Ops Layer
This project uses the Convergent Methods ops layer protocol. The canonical protocol definition lives in /Users/willwright/ConvergentMethods/_strategy/OPS_LAYER_PROTOCOL.md. Read it at session boot. Follow all ops layer behaviors defined there: session boot sequence, session end protocol, amendment protocol, session log rotation, status bar, and directory scope rules.
On session boot, read ROADMAP.md and SESSION_LOG.md at the repo root before doing any work. These files govern phase sequencing and session continuity.
Upward propagation paths: When a propagation event occurs (phase completion, external publish, blocking status change, HITL gate reached/cleared — see CM root CLAUDE.md for full list), update these CEO-level docs before the session ends:
/Users/willwright/ConvergentMethods/MASTER.md— this project's workstream entry/Users/willwright/ConvergentMethods/OPUS_BRIEF.md— this project's portfolio brief entry
Status bar: Every response footer must include, alongside existing lines (docs updated, persistence receipts, timestamp):
Project: Boyce | Phase: [current phase from ROADMAP.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.
- 3d ago First seen · 489 lines · 6,243 tokens per session scan A a647887b1969
boyce CLAUDE.md is an instructions file published in the GitHub repository boyce-io/boyce (1 stars, last pushed 4mo ago), licensed MIT. It adds 6,243 tokens to every session, about $0.0312 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
orionbelt-semantic-layer-mcp AGENTS.md
AGENTS.md instructions for ralforion/orionbelt-semantic-layer-mcp, covering agents.md, what this is, commands, architecture (the non-obvious parts) and version tracking & release discipline.
orionbelt-semantic-layer-mcp CLAUDE.md
Claude Code instructions for ralforion/orionbelt-semantic-layer-mcp, a project described as: MCP server for the OrionBelt Semantic Layer. Enables LLMs to explore semantic models, compile queries, and execute analytics via natural language. Works with Claude, Cursor, Windsurf, Copilot.
turso AGENTS.md
Instructions for tursodatabase/turso, covering turso agent guidelines, quick reference, testing, running tests and test organization.
sqlglot AGENTS.md
Instructions for tobymao/sqlglot, covering contributing to sqlglot, about sqlglot, development commands, installation and basic installation.
spiceai copilot-instructions.md
Copilot instructions for spiceai/spiceai, covering spice.ai agent instructions, data correctness — absolute top priority, evidence — no claim without a reproduction, build, test, lint (expensive — read first) and git & prs.
graphjin AGENTS.md
AGENTS.md instructions for dosco/graphjin, covering graphjin agent guide, architectural overview, directory structure & responsibilities, build commands and coding guidelines.