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/instructa/codetie/001-basegit clone --depth 1 https://github.com/instructa/codetieWhat 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 | $0.00376 | $0.00376 |
| Opus 5 | $0.00188 | $0.00188 |
| Sonnet 5 | $0.00075 | $0.00075 |
| Haiku 4.5 | $0.00038 | $0.00038 |
Grade A, and why
001-base 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.
What it actually says
Instructions
- Always use codebase_search with target_directories="src,app" first to find existing core files
- Always check existing system files' purposes before creating new ones with similar functionality
- Always list the cursor rules you're using
- If a prompt or request specifies certain behaviors, languages, or output formats, you must obey them without deviation.
- Do not include explanations, reasoning, or filler text unless explicitly instructed. Stick strictly to the requested output.
- If multiple steps or sub-requests are given, address them in the specified order. Provide answers in the exact format or sequence requested.
- Pay close attention to all stated constraints (e.g., language choice, performance goals, coding style). Do not ignore any requirement or best practice stated.
- Only produce output relevant to the question or instructions. Do not add features, code, or details beyond what is explicitly asked.
- Deliver the response in a minimal yet complete form. Avoid unnecessary verbosity and tangential remarks.
- If the prompt requests a specific output format (e.g., a fenced code block, bullet points, JSON), follow that format exactly.
- If a prompt includes a pre-seeded answer structure (e.g., starts a code block), continue within that structure without introducing extra text outside it.
- If the request is ambiguous, you may ask clarifying questions (if instructions allow). Otherwise, state briefly that more information is needed.
- When generating or modifying code, adhere to best practices for clarity, maintainability, and efficiency, as appropriate to the specified language or framework.
- Do not generate or include private data (API keys, secrets) unless explicitly provided in context. If the user requests something unsafe or disallowed, refuse or provide a safe alternative per policy.
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 · 24 lines · 376 tokens per session scan A 8e306a3005c4
001-base is a cursor rule published in the GitHub repository instructa/codetie (8 stars, last pushed 1y ago), licensed MIT. It adds 376 tokens to every session, about $0.0019 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 cursor rules, from other repositories
angular
Angular: signals, standalone components, RxJS patterns.
django
Django: models, views, ORM best practices.
java
Modern Java: records, sealed classes, streams, virtual threads.
javascript
Modern JavaScript: ES2023+, async patterns, common traps.
rust
Rust patterns: ownership, Result types, iterators.
accessibility
Accessibility: semantic HTML, ARIA, keyboard navigation, testing.