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/jeremyrayner/kb-template/claude-mdgit clone --depth 1 https://github.com/jeremyrayner/kb-templateWrote 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/jeremyrayner/kb-template/claude-md)<a href="https://agentmods.dev/instructions/jeremyrayner/kb-template/claude-md"><img src="https://agentmods.dev/badge/instructions/jeremyrayner/kb-template/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 5.1 | $0.01390 | $0.01390 |
| Opus 5 | $0.00695 | $0.00695 |
| Sonnet 5 | $0.00278 | $0.00278 |
| Haiku 4.5 | $0.00139 | $0.00139 |
Grade A, and why
kb-template 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 5d 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 — 128 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge Base — Schema
Portability note: This file is named
CLAUDE.mdfor Claude Code. Rename it to match your LLM agent's convention (e.g.AGENTS.mdfor Codex). The instructions are AI-agnostic.
This is an LLM-maintained knowledge base. The LLM writes and maintains all wiki content. The human curates sources, directs analysis, and asks questions. All data is stored as plain files in universal formats (markdown, images) — the human owns the data and can use any AI agent or tool over it.
Architecture
Three layers:
raw/— Immutable source documents. Articles, papers, images, data files. The LLM reads from these but never modifies them. This is the source of truth.wiki/— LLM-generated and LLM-maintained markdown files. Summaries, entity pages, concept pages, comparisons, synthesis. The LLM owns this layer entirely.output/— Generated artifacts (reports, slide decks, charts). These are outputs of Q&A and analysis operations.
Wiki conventions
Page format
Every wiki page must have YAML frontmatter:
---
tags: [topic-a, topic-b]
date: YYYY-MM-DD
sources: 3
---
tags: relevant topic tagsdate: creation date of the pagesources: count of raw sources that contributed to this page
Links
Use [[wikilinks]] for all internal links. Never use [text](path.md) style.
Page structure
Each page should follow this pattern:
---
frontmatter
---
# Page Title
Main content with [[wikilinks]] to related concepts.
## Sources
- [[raw/articles/source-file]]
Special files
wiki/index.md— Content catalog. Every wiki page listed with a link, one-line summary, and metadata. Organised by category. The LLM reads this first when answering queries.wiki/overview.md— Top-level synthesis page. A narrative that ties together the main themes, key findings, open questions, and tensions across all sources. Not a table of contents — that's the index. Update this whenever an ingest shifts the big picture.wiki/log.md— Append-only chronological record of operations. Each entry uses the format## [YYYY-MM-DD] type | Titlewhere type is one of:ingest,query,lint,update. This format is grep-friendly:grep "^## \[" wiki/log.md | tail -5.
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.
- 5d ago First seen · 128 lines · 1,390 tokens per session scan A 6243e4e242b5
kb-template CLAUDE.md is an instructions file published in the GitHub repository jeremyrayner/kb-template (24 stars, last pushed 5mo ago), licensed MIT. It adds 1,390 tokens to every session, about $0.0069 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.
Other instructions, from other repositories
lms-front CLAUDE.md
Claude Code instructions for guillermoscript/lms-front, covering claude.md, project overview, commands, architecture and multi-tenancy.
vespertide AGENTS.md
AGENTS.md instructions for dev-five-git/vespertide, covering vespertide knowledge base, structure, where to look, data flow and conventions.
seekstone CLAUDE.md
Claude Code instructions for shaqmughal/seekstone, covering claude.md, what this repo is, commands, the harness itself (run after npm install) and architecture.
rails_ai_agents AGENTS.md
AGENTS.md instructions for ThibautBaissac/rails_ai_agents, covering project configuration, tech stack, architecture, key commands and tests.
pg-dash CLAUDE.md
Claude Code instructions for indiekitai/pg-dash, covering pg-dash claude.md, 项目结构, 构建与测试, 发版纪律(强制) and 发版检查清单(每次 npm publish 前必须按序执行).
MCP-SqlServer CLAUDE.md
Claude Code instructions for Aron-Valenzuela/MCP-SqlServer, covering claude.md, project overview, setup (for each user), 1. install dependencies and 2. configure claude desktop.