Borrowing it
Nothing to install: this file belongs to tanghong123/claude-replay. 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/tanghong123/claude-replay/main/CLAUDE.mdgit clone --depth 1 https://github.com/tanghong123/claude-replayWrote 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/tanghong123/claude-replay/claude-md)<a href="https://agentmods.dev/instructions/tanghong123/claude-replay/claude-md"><img src="https://agentmods.dev/badge/instructions/tanghong123/claude-replay/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/tanghong123/claude-replay/claude-md"><img src="https://agentmods.dev/badge/instructions/tanghong123/claude-replay/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 | $0.06503 | $0.06503 |
| Opus 5 | $0.03252 | $0.03252 |
| Sonnet 5 | $0.01301 | $0.01301 |
| Haiku 4.5 | $0.00650 | $0.00650 |
Grade A, and why
claude-replay 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 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 — 329 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md — claude-replay
A Rust + ratatui terminal UI viewer (binaries: agent-replay / agent-monitor /
agent-monitor-fleet since v1.101.0 — crate and repo names keep the claude- prefix;
brew installs symlink the old command names). It is fully testable headless (no TTY) —
never skip, stub, or defer a feature "because it needs a terminal."
Work tracking
The tasks/ queue is the state of record for pending work (owner, 2026-08-30; it
replaced BACKLOG.md, now a pointer). Design docs argue, issues discuss, the queue tracks
— don't trust a design/*.md status header alone, since the tracker exists because those
drift. How the queue is driven is a local-agent concern and lives in the machine-level
CLAUDE.md, not here.
Two habits this repo cares about, because it is the thing that RENDERS the result. Both are unrecoverable after the fact, and both were learned from real damage here:
- Never pipe or redirect a mutation.
taskqprints its##taskq/v1record as the last line of stdout, and that record is how the work reaches the task panel of every agent's transcript.| tail,| head,> /dev/nulland2>&1 |all destroy it. Measured on one session: 17donecommands, every one piped, 6 records surviving — and 12 tasks still rendering as pending when the owner looked. v1.128.0 recovers much of it from the command line and from taskq's own prose, but recovery is guesswork where the record was fact. - Pass
--descriptionas a literal, never"$D". The shell expands the variable before taskq sees it, so the task file is right and the TRANSCRIPT holds$D. Multi-paragraph text in single quotes is fine; a heredoc into a variable is the trap.
The monitor's two shells
agent-monitor and agent-monitor-v2 each serve two frontends and both are supported: the
app shell (claude-monitor/src/ui.rs + src/codex-ui/*, the default) and the classic
page (v1's rail, v2's splice shell). A button in each switches and REMEMBERS the choice at
<state_dir>/ui.json, shared by both binaries; ?ui=classic / ?ui=app override for one
request without disturbing it, which is what makes side-by-side comparison possible. The classic
page is not deprecated — it goes when the app shell has been validated, and not before.
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 Changed · +9 lines · +176 tokens per session b07d1c419ade
- 4d ago Changed · +6 lines · +127 tokens per session 0061a571b970
- 6d ago Changed · +89 lines · +1,952 tokens per session 86267ab8fb2d
- 10d ago First seen · 225 lines · 4,248 tokens per session scan A 59b81c1c6cce
claude-replay CLAUDE.md is an instructions file published in the GitHub repository tanghong123/claude-replay (2 stars, last pushed yesterday), licensed MIT. It adds 6,503 tokens to every session, about $0.0325 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
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.
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
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).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
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.