Borrowing it
Nothing to install: this file belongs to 0xsaju/claude-standby. 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/0xsaju/claude-standby/main/CLAUDE.mdgit clone --depth 1 https://github.com/0xsaju/claude-standbyWrote 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/0xsaju/claude-standby/claude-md)<a href="https://agentmods.dev/instructions/0xsaju/claude-standby/claude-md"><img src="https://agentmods.dev/badge/instructions/0xsaju/claude-standby/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/0xsaju/claude-standby/claude-md"><img src="https://agentmods.dev/badge/instructions/0xsaju/claude-standby/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 | $0.01366 | $0.01366 |
| Opus 5 | $0.00683 | $0.00683 |
| Sonnet 5 | $0.00273 | $0.00273 |
| Haiku 4.5 | $0.00137 | $0.00137 |
Grade B, and why
claude-standby CLAUDE.md scanned grade B with 2 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 9d 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.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
~/.claude/settings.json (opt-in; python3 JSON edit, backup, reversible) Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- `install.sh` — curl-pipe-bash installer; links the CLI (no hooks, no plugin) How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
claude-standby — guardrails
Tool that lets Claude Code tasks survive rate-limit hits: detect the limit,
wait until reset, auto-resume the session. Engine = portable scripts
(plugin/scripts/), cockpit = VS Code extension (vscode-extension/),
contract = ~/.claude/auto-resume/state.json. The Claude Code Stop-hook
path was removed 2026-07-19 (D31): detection reads local data, not hooks.
Hard constraints — do not violate
- C1 — No invented data shapes. Positive detection — the decision that a
limit is active, or that it has lifted — may only rely on formats
MEASURED and documented in
docs/HOOK-FINDINGS.md: F1 (the limit MESSAGE the probe reads), F2 (the session store for--resumeids), F4 (the status-line rate stream:used_percentage+resets_at). No guessed payloads. Auto-detect = read a rate snapshot (F4) if present, else onehaikuprobe that trusts the F1 message — never the exit code. Anything outside F1/F2/F4 (an alternate cache field such asrate_pct, opt-in stream-json output, a loosely-shaped session id) is informational/opt-in only: it may be displayed or tried, but it never by itself authorizes a resume, and the F1 bounce guard still re-checks the real outcome before a resume counts as successful. - C2 — Portable bash. POSIX-compatible bash, no hard
jqdependency (fallback chain: jq → python3 → awk/sed on canonical layout), no GNU-only flags without a BSD alternative. Target Linux + macOS; Windows best-effort. - C3 — Engine layout. The engine is plain scripts under
plugin/scripts/(theplugin/name is legacy — there is no Claude Code plugin anymore; the plugin manifest + marketplace were removed with the hooks, D33). The CLI (bin/claude-standby) is the only interface (D17); version lives in the top-levelVERSIONfile. No slash commands (they cost tokens and can't run while limited). - C4 — Sensors never break the host. The status-line sensor
(
statusline.sh) alwaysexit 0, finishes fast, chains (never clobbers) any existing status line, never stderr noise. It must be invisible when it fails. - C5 — Safety rails.
max_resumesenforced (numeric-validated, fails closed on garbage); a conservative default permission allowlist on every unattended resume unless the user supplies their ownEXTRA_ARGS(--dangerously-skip-permissionsonly behind explicit opt-in, never added by default); optional quiet hours (off by default, deferral only — never resumes early); progress-stall / clean-but-idle outcome detection marks a taskstuckinstead of silentlydonewhen a workspace has a progress file to compare against. - C6 — Real quota is precious. All iterative testing uses
test/fake-claude.sh. Real limit burns only for milestone verification.
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.
- 9d ago First seen · 87 lines · 1,366 tokens per session scan B eb234e4c6232
claude-standby CLAUDE.md is an instructions file published in the GitHub repository 0xsaju/claude-standby (5 stars, last pushed 1mo ago), licensed MIT. It adds 1,366 tokens to every session, about $0.0068 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). 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).
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.
deepseek-harness AGENTS.md
AGENTS.md instructions for deepseek-ai/deepseek-harness, covering agents.md, pre-stable apis and released session data, repository layout, commands and host sandbox failures.