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 rules/littlebearapps/untether/cursorrulesgit clone --depth 1 https://github.com/littlebearapps/untetherWhat 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 | $0.00487 | $0.00487 |
| Opus 5 | $0.00244 | $0.00244 |
| Sonnet 5 | $0.00097 | $0.00097 |
| Haiku 4.5 | $0.00049 | $0.00049 |
Grade A, and why
cursorrules 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 — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Untether — Cursor Rules
Telegram bridge for AI coding agents (Claude Code, Codex, OpenCode, Pi, Gemini CLI, Amp). Lets users control agents from their phone or any device — agents run on the user's machine while they're away from the terminal.
Stack & conventions
- Python 3.12+, anyio for async, msgspec for JSONL parsing, structlog for logging
- Ruff for linting/formatting, pytest + anyio for testing (80% coverage threshold)
- Australian English in user-facing text (realise, colour, behaviour, licence)
- Conventional commits: feat:, fix:, docs:, refactor:, test:
Architecture
Telegram <-> TelegramPresenter <-> RunnerBridge <-> Runner
|
ProgressTracker
- Runners (src/untether/runners/) — engine subprocess managers
- RunnerBridge (src/untether/runner_bridge.py) — connects runners to transport
- TelegramPresenter (src/untether/telegram/bridge.py) — message rendering
- Commands (src/untether/telegram/commands/) — Telegram command handlers
- Schemas (src/untether/schemas/) — msgspec structs for JSONL decoding
Key rules
- Runner 3-event contract: StartedEvent → ActionEvent(s) → CompletedEvent (always)
- Use EventFactory for event construction, never construct dataclasses directly
- ALL Telegram writes go through TelegramOutbox (never call Bot API directly)
- Callback data max 64 bytes (Telegram-enforced)
- Use anyio (not raw asyncio), msgspec.Struct (not dataclasses for schemas)
Testing
- Stub subprocess runners with fake CLI scripts
- FakeTransport protocol doubles (not real Telegram clients)
- Verify 3-event contract in runner tests
- Run: uv run pytest, uv run ruff check src/
Key files
- runners/claude.py — Claude Code runner with interactive features
- runner_bridge.py — Runner-to-transport bridge
- cost_tracker.py — Per-run/daily cost tracking
- telegram/bridge.py — Telegram message rendering
- commands/claude_control.py — Approve/Deny/Discuss callback handler
- commands/stats.py — /stats session statistics
- commands/auth.py — /auth Codex device re-authentication
- markdown.py — Progress/final message formatting
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 · 51 lines · 487 tokens per session scan A 80daade9bdf4
cursorrules is a cursor rule published in the GitHub repository littlebearapps/untether (66 stars, last pushed 2d ago), licensed MIT. It adds 487 tokens to every session, about $0.0024 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 cursor rules, from other repositories
architecture-constraints
GolemBot architecture hard constraints — must check before modifying any src/ code.
cursor-memory
Shokunin memory system: persistent context across sessions.
skill-conventions
Project conventions for creating and editing Obsidian vault skills.
notes
The .cursor/notes folder contains notes about this project. The notes form a knowledge base about the project to be used by the AI Agents working on the project. They should help you when working with the same code again in the future.
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.