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 skills/vindm/dotclaude/knowledge-layersnpx skills add vindm/dotclaude --skill knowledge-layersgit clone --depth 1 https://github.com/vindm/dotclaudeWhat 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 | $0.00089 | $0.00952 |
| Opus 5 | $0.00044 | $0.00476 |
| Sonnet 5 | $0.00018 | $0.00190 |
| Haiku 4.5 | $0.00009 | $0.00095 |
Grade A, and why
knowledge-layers 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 yesterday.
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 — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge layers — authority order
A project has three knowledge layers, and they rank in a fixed authority order. Most doc-trust bugs come from not knowing which one wins when they disagree.
-
.claude/(or the project's guidance layer) is GUIDANCE. Rules, skills, invariants, conventions — how to build this project. Intentional, hand-authored, prescriptive. It tells you what discipline to apply; it does not describe what the system currently does. -
Code is TRUTH. It is what the system actually does, because it is the thing that runs. For any factual question about real behavior — the actual retry limit, the actual auth check, the actual schema — read code, not a description of code.
-
docs/is a REFLECTION of code, plus transient intent. Reflection docs (architecture overviews, flow docs, a capabilities index) are a downstream description of code — a navigable map to find the right code fast, not an authority on what it does. Transient intent (brainstorms, plans) is forward-looking exploration consumed when the work ships, then archived. Neither is authoritative about current behavior.
The rule that follows from the ranking
Derive correctness from code plus the guidance layer. Verify any load-bearing claim against code. When a doc and the code disagree, the code wins — every time, no exceptions. This is not "prefer code"; it is code wins. Hedged language invites you to trust the doc anyway.
The reason: code is executed, so it can't lie for long; a doc is never executed, so it can be wrong indefinitely and look fine. docs/ is downstream of code, so it drifts, and drift is silent. Pointing authority at code is how you stop trusting the layer that rots.
Three consequences for how you work
- Code → docs, never the reverse. When you change code that a reflection doc describes, update that doc in the same change, or stale-mark it with a "verify against code" note. Correctness flows one direction: code is the source, docs follow. Never edit a reflection doc to declare new behavior and expect code to catch up — that inverts the authority order and manufactures drift. (Specs and plans are transient intent, and may legitimately lead code; a reflection doc must never lead it.)
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.
- yesterday First seen · 36 lines · 89 tokens per session scan A dc79989a49cd
knowledge-layers is a skill published in the GitHub repository vindm/dotclaude (1 stars, last pushed 5d ago), licensed MIT. It adds 89 tokens to every session and 952 once invoked, about $0.0004 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 skills, from other repositories
claude-md-review
Audit a CLAUDE.md file for the patterns that actually degrade Claude Code's output — vagueness, unnamed files, stale facts, and bloat. Use when asked to review, audit, improve, shrink, or fix a CLAUDE.md, and when a project's results feel inconsistent or Claude keeps rediscovering the same context.
git-commit-helper
Generate conventional commit messages automatically. Use when user runs git commit, stages changes, or asks for commit message help. Analyzes git diff to create clear, descriptive conventional commit messages. Triggers on git commit, staged changes, commit message requests.
code-reviewer
Automatic code quality and best practices analysis. Use proactively when files are modified, saved, or committed. Analyzes code style, patterns, potential bugs, and security basics. Triggers on file changes, git diff, code edits, quality mentions.
test-generator
Automatically suggest tests for new functions and components. Use when new code is written, functions added, or user mentions testing. Creates test scaffolding with Jest, Vitest, Pytest patterns. Triggers on new functions, components, test requests, testing mentions.
api-documenter
Auto-generate API documentation from code and comments. Use when API endpoints change, or user mentions API docs. Creates OpenAPI/Swagger specs from code. Triggers on API file changes, documentation requests, endpoint additions.
readme-updater
Keep README files current with project changes. Use when project structure changes, features added, or setup instructions modified. Suggests README updates based on code changes. Triggers on significant project changes, new features, dependency changes.