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.
git clone --depth 1 https://github.com/zotoio/CRUX-CompressWrote 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/rules/zotoio/crux-compress/skill-and-agent-references)<a href="https://agentmods.dev/rules/zotoio/crux-compress/skill-and-agent-references"><img src="https://agentmods.dev/badge/rules/zotoio/crux-compress/skill-and-agent-references/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/rules/zotoio/crux-compress/skill-and-agent-references"><img src="https://agentmods.dev/badge/rules/zotoio/crux-compress/skill-and-agent-references.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 | $0.01439 | $0.01439 |
| Opus 5 | $0.00720 | $0.00720 |
| Sonnet 5 | $0.00288 | $0.00288 |
| Haiku 4.5 | $0.00144 | $0.00144 |
Grade A, and why
skill-and-agent-references 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 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.
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 — 94 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill and Agent References
Skills, agents, and commands in Cursor are registered with the IDE/CLI by name — the directory name under .cursor/skills/<name>/SKILL.md, the file basename under .cursor/agents/<name>.md, and the file basename under .cursor/commands/<name>.md, mirrored in the name: frontmatter field where present. The IDE skill loader, the Task subagent dispatcher, the slash-command surface, and the skill resolution machinery all look up a primitive from this registered name. Filesystem paths are an implementation detail.
This rule keeps every operational reference stable, portable, and resilient to file moves.
Rules
1. Reference skills, agents, and commands by name only
Always write the registered name in backticks:
- Skill:
`crux-skill-memory-extract` - Agent:
`crux-cursor-meditation-guide` - Command:
/crux-meditate
Never embed the filesystem path in instructional or operational prose. Examples to avoid:
Bad: `.cursor/skills/crux-skill-memory-extract/SKILL.md`
Bad: [`.cursor/skills/foo/SKILL.md`](../skills/foo/SKILL.md)
Bad: Read `.cursor/agents/crux-cursor-memory-manager.md` to see how it works.
Bad: Load .cursor/skills/crux-skill-memory-meditation-research/SKILL.md for Research mode.
Good: Use the `crux-skill-memory-extract` skill.
Good: Spawn a `crux-cursor-memory-manager` subagent.
Good: Load the `crux-skill-memory-meditation-research` skill for Research mode.
The Cursor skill loader and the Task tool resolve these names automatically. Paths break the moment a file moves, double the maintenance cost (path + name), add visual noise, and falsely imply the file must be opened directly when it should be loaded through the standard skill or agent mechanism.
This applies to:
- Commands under
.cursor/commands/— refer to as/<command-name> - Agents under
.cursor/agents/— refer to by<agent-name>in backticks - Skills under
.cursor/skills/<name>/SKILL.md— refer to by<skill-name>in backticks - Any other Cursor primitive that is loaded by name
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 · 94 lines · 1,439 tokens per session scan A a4b2510bd526
skill-and-agent-references is a cursor rule published in the GitHub repository zotoio/CRUX-Compress (8 stars, last pushed 2d ago), licensed MIT. It adds 1,439 tokens to every session, about $0.0072 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
swift6-migration
Guide for migrating Swift code to Swift 6 with concurrency and strict checking.
swift-argument-parser
Documentation and usage patterns for Swift Argument Parser for command-line tools.
swift-testing-playbook
Comprehensive guide for migrating from XCTest to Swift Testing with best practices from WWDC 2024.
swift-observation
Swift Observation framework documentation and usage patterns for reactive programming.
swift-observable
Migration guide for updating SwiftUI apps from ObservableObject to the @Observable macro.
screenshot-automation
Reusable AppleScript patterns for automated screenshot capture for documentation and testing.