03-shared

Shared project rules for TypeScript, translation, pull requests, development sessions, and the team's terminology.

In plain words
What is it for?
Use them to follow project-wide coding and review conventions, track session progress, and apply the team's quality and delivery vocabulary.
Why use it?
They give contributors common ways to describe work and handle cross-cutting development practices.

Cursor rule

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.

agentmods
npx agentmods add rules/wellapp-ai/well/03-shared
Clone the repo
git clone --depth 1 https://github.com/WellApp-ai/Well
Per session 2,197 This file is loaded in full into every session.
When invoked 2,197 The same file — it is already loaded in full.
Security scan A 0 findings. Scan, not verified.
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 $0.02197 $0.02197
Opus 5 $0.01099 $0.01099
Sonnet 5 $0.00439 $0.00439
Haiku 4.5 $0.00220 $0.00220

Measured yesterday against content hash 658f296e3d3d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

03-shared 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.

cursor-rules/rules/03-shared.mdc · 203 lines

How it starts

The opening of the file, as written. The whole thing — 203 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Shared Rules

Ubiquitous Vocabulary

Term Definition
Value Analysis Determine WHAT to build (Ask Mode)
Value Engineering Determine HOW to build (Plan Mode)
Value Delivery Build, verify, ship (Agent/Commit/PR Modes)
QA Contract G#N Gherkin scenarios + AC#N acceptance criteria
Gate Human decision checkpoint (Gates 1-4)
Commit Atomic unit of shippable value
Patine Accumulated wisdom of why decisions were made
BLUF Bottom Line Up Front - answer/decision in first sentence
Progressive Disclosure L1 summary default, L2/L3 detail on request

Session Tracking

Term Definition
Session Development work from init to PR push
Loop One iteration within a phase (L1, L2...)
Status Flow state: G (green/flow), Y (yellow/waiting), R (red/stopped)
Breadcrumb Header + footer showing session position
Session Journal Aggregated metrics + Patine per session (Notion)

TPS (Toyota Production System)

Term Plain Definition
Takt Pace Target time per phase with warning/stop thresholds
Muda Waste Rework count, waiting time, over-processing
Jidoka Escalation Stop when stuck; Tier 1 (auto), Tier 2/3 (human)
Kaizen Improvement Patterns captured from repetition
Hansei Reflection Learnings from corrections and timeouts
Poka-Yoke Validation Silent pre-Gate artifact checks
Autonomous Hands-off AI iterates until success or escalation

Action Keywords

Keyword Effect
OK Approve current output, proceed
KO Reject, stop and reassess
DIG Request more depth, increment Loop
continue Next step in current phase
plan Transition to PLAN mode
agent Transition to AGENT mode
push Transition to PUSH-PR mode
ralph mode Trigger Autonomous Mode
pause and show options Override auto-proceed, show Safe/Balanced/Bold

Read the full file on GitHub · 203 lines

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. yesterday First seen · 203 lines · 2,197 tokens per session scan A 658f296e3d3d

Subscribe to this mod's changes

03-shared is a cursor rule published in the GitHub repository WellApp-ai/Well (341 stars, last pushed 25d ago), licensed MIT. It adds 2,197 tokens to every session, about $0.0110 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.