03-collaboration-modes

03-collaboration-modes is a cursor rule for Cursor from alieismy/codex-three-layer-delivery. It costs 360 tokens per session, scanned A, original, MIT.

A set of rules for deciding how people and AI should share work on document projects, based on risk, uncertainty, decision authority, and the strength of available checks.

In plain words
What is it for?
Use it to choose between mostly AI drafting, mostly human decision-making, or shared work for requirements, feasibility studies, designs, and reviews.
Why use it?
It helps prevent an AI from making unsupported business, compliance, standards, or architecture decisions when a person needs to decide.

Cursor rule for Cursor

Written for Cursor: installed under .cursor/.

Good fit Use it to choose between mostly AI drafting, mostly human decision-making, or shared work for requirements, feasibility studies, designs, and reviews.

Compare 6 cursor rules from other repositories ↓
Install with agentmods
npx agentmods add rules/alieismy/codex-three-layer-delivery/03-collaboration-modes
Install

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.

Clone the repo
git clone --depth 1 https://github.com/alieismy/codex-three-layer-delivery

Made for: Cursor.

Wrote 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.

agentmods badge for 03-collaboration-modes

README.md
[![agentmods](https://agentmods.dev/badge/rules/alieismy/codex-three-layer-delivery/03-collaboration-modes.svg)](https://agentmods.dev/rules/alieismy/codex-three-layer-delivery/03-collaboration-modes)
Your own site
<a href="https://agentmods.dev/rules/alieismy/codex-three-layer-delivery/03-collaboration-modes"><img src="https://agentmods.dev/badge/rules/alieismy/codex-three-layer-delivery/03-collaboration-modes.svg" alt="Measured on agentmods" height="20"></a>
Per session 360 This file is loaded in full into every session.
When invoked 360 The same file — it is already loaded in full.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00360 $0.00360
Opus 5 $0.00180 $0.00180
Sonnet 5 $0.00072 $0.00072
Haiku 4.5 $0.00036 $0.00036

Measured 7d ago against content hash 8881bb0ecb41, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

03-collaboration-modes 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 7d 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.

cursor/project/.cursor/rules/03-collaboration-modes.mdc · 42 lines

What it actually says

Human-AI Collaboration Modes

Select the collaboration mode based on document risk, ambiguity, authority, and verification strength.

Mode Use when Human effort Agent effort
AI First Template-based PRD sections, review tables, glossary cleanup, structured summaries 20% review 80% drafting
Human First Business scope decisions, architecture trade-offs, standards authority, approval conclusions 70% decision 30% analysis
Pair Mode Feasibility analysis, proposal comparison, detailed design, clause-level standards work 50% 50%

Default to Pair Mode when the work has business, compliance, standards, or architecture consequences.

AI First Preconditions

  • Scope is clear and bounded.
  • Source material and existing templates are available.
  • Output format and acceptance criteria are clear.
  • Verification commands or document consistency checks exist.

Human First Triggers

  • Business scope, approval responsibility, or standards authority is unresolved.
  • Architecture boundaries or compliance constraints are disputed.
  • Evidence is insufficient for a factual, regulatory, or standards claim.
  • The decision could become an authoritative baseline.

Pair Mode Behavior

  • Read and summarize relevant documents first.
  • Propose a short plan with verification points for multi-document work.
  • Draft or revise in narrow increments.
  • Verify traceability and evidence after each meaningful step when practical.

Background or Parallel Work

Use background agents or parallel tasks only when the subtasks are isolated and can be verified independently. Keep write access scoped to the document sections or files required for the task.

Changes

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.

  1. 7d ago First seen · 42 lines · 360 tokens per session scan A 8881bb0ecb41

Subscribe to this mod's changes

03-collaboration-modes is a cursor rule published in the GitHub repository alieismy/codex-three-layer-delivery (2 stars, last pushed 7d ago), licensed MIT. It adds 360 tokens to every session, about $0.0018 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.