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/microsoft/clarity/claude-mdgit clone --depth 1 https://github.com/microsoft/clarityWhat 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 | $0.01626 | $0.01626 |
| Opus 5 | $0.00813 | $0.00813 |
| Sonnet 5 | $0.00325 | $0.00325 |
| Haiku 4.5 | $0.00163 | $0.00163 |
Grade A, and why
clarity 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.
This is a copy
94% identical to clarity copilot-instructions.md — 4 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 143 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Clarity Development Guidelines
Repository Overview
Clarity is an open-source TypeScript behavioral analytics library for tracking user interactions and session replays. Monorepo with Lerna/Yarn workspaces:
- clarity-js (packages/clarity-js): Core instrumentation library
- clarity-decode (packages/clarity-decode): Data decoder
- clarity-visualize (packages/clarity-visualize): Session replay visualization
- clarity-devtools (packages/clarity-devtools): Chrome extension (private)
Stack: TypeScript, Rollup, TSLint, Playwright, Lerna, Yarn workspaces
clarity-js Performance & Bundle Size Priorities
Critical: clarity-js has two top priorities that must guide all development decisions:
-
Performance - As an analytics library running on diverse websites, clarity-js must not hurt website performance or impact metrics like INP (Interaction to Next Paint) or PLT (Page Load Time). Balance data collection needs with:
- Avoid blocking the main thread
- Minimize network requests
- Reduce payload size
-
Bundle Size - Small bundle size is crucial for loading clarity-js quickly and starting data collection as soon as possible. Every byte matters.
Build & Development Commands
Installation
ALWAYS run yarn install before building or testing. (~25s, peer dependency warnings are normal)
Building
yarn build # All packages (~45s, Lerna parallel build)
yarn build:js # clarity-js only (~20s)
yarn build:decode # clarity-decode only
yarn build:visualize # clarity-visualize only
yarn build:devtools # clarity-devtools only
Rollup + TypeScript → multiple formats (CJS, ESM, IIFE minified). Build artifacts auto-cleaned to build/ or extension/ (gitignored).
Build outputs per package:
- clarity-js:
clarity.js(CJS),clarity.module.js(ESM),clarity.min.js(minified), plus variants:clarity.extended.js,clarity.insight.js,clarity.performance.js,clarity.livechat.js, and dynamic module builds (clarity.tidio.js,clarity.crisp.js) - clarity-decode:
clarity.decode.js(CJS),clarity.decode.module.js(ESM),clarity.decode.min.js(minified) - clarity-visualize:
clarity.visualize.js(CJS),clarity.visualize.module.js(ESM),clarity.visualize.min.js(minified) - clarity-devtools: Output to
extension/directory for Chrome extension
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 · 143 lines · 1,626 tokens per session scan A 436fa64d6262
clarity CLAUDE.md is an instructions file published in the GitHub repository microsoft/clarity (2,729 stars, last pushed today), licensed MIT. It adds 1,626 tokens to every session, about $0.0081 per session on Opus 5. A static security scan graded it A with 0 findings. It is 94% identical to clarity copilot-instructions.md, differing in 4 lines, and is treated as a copy.
Other instructions, from other repositories
lastsearch CLAUDE.md
Claude Code instructions for LastSearch-HQ/lastsearch, covering claude.md — lastsearch, what this project is, monorepo structure, two-repo architecture and key commands.
opslane AGENTS.md
AGENTS.md instructions for opslane/opslane, covering agents.md, areas, verification, cross-cutting conventions and guardrails.
thermal AGENTS.md
AGENTS.md instructions for jadmadi/thermal, covering agents.md — thermal coding standards & architecture guide, 1. project purpose & architecture, directory structure, 2. core coding & architectural standards and a. read-only data ingestion (internal/loaders/).
codex-usage-heatmap AGENTS.md
AGENTS.md instructions for Gatsby007max/codex-usage-heatmap, covering codex usage heatmap agent guide, project overview, repository layout, development commands and build, test, and lint.
opslane CLAUDE.md
Claude Code instructions for opslane/opslane, a project described as: Opslane finds the bugs your users run into by watching real user sessions, and writes the fix. You review and merge.
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).