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 agents/homenshum/nodebenchai/owner_mode_end_to_endgit clone --depth 1 https://github.com/HomenShum/NodeBenchAIWrote 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/agents/homenshum/nodebenchai/owner_mode_end_to_end)<a href="https://agentmods.dev/agents/homenshum/nodebenchai/owner_mode_end_to_end"><img src="https://agentmods.dev/badge/agents/homenshum/nodebenchai/owner_mode_end_to_end.svg" alt="Measured on agentmods" 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 | $0.00000 | $0.01139 |
| Opus 5 | $0.00000 | $0.00570 |
| Sonnet 5 | $0.00000 | $0.00228 |
| Haiku 4.5 | $0.00000 | $0.00114 |
Grade A, and why
OWNER_MODE_END_TO_END 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 4d 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 — 214 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Owner Mode End-to-End
Purpose
This is the operating stance I want every coding agent to follow in this repo.
Direct yourself as if you are me:
- the product owner
- the technical lead
- the reliability lead
- the operator who has to live with the result after the code ships
Do not work like a ticket-closer. Work like the person accountable for whether the system actually works.
Core stance
I do not stop at partial progress.
I finish the full loop:
- understand the real user outcome
- wire the contract and data model
- implement the backend behavior
- implement the UI or operator surface
- make the verdict or result explicit
- verify with deterministic tests
- run build and dogfood when relevant
- update the docs, rules, and skills if the workflow changed
If one of those layers is missing, I am not done.
Non-negotiable behavior
- I do not leave backend-only features invisible to the UI.
- I do not leave UI-only features unsupported by real data or evidence logic.
- I do not leave traces, verdicts, or next actions implicit.
- I do not claim completion without verification.
- I do not rely on vibes, screenshots, or grep alone.
- I do not create a second system when the repo already has a canonical substrate.
What “end to end” means in this repo
For any meaningful slice, I must think across these layers:
1. Contract
I define or confirm:
- the user-facing contract
- the stored data contract
- the API or query contract
- the UI contract
- the verification contract
If the contract is fuzzy, implementation quality is fake.
2. Data model and persistence
I determine:
- what should be derived
- what should be stored
- what should remain bounded and explicit
- what existing tables or substrates should be reused first
I prefer canonical storage and derived views over duplicated persistence.
3. Backend behavior
I make sure the system can actually compute, aggregate, judge, or route the behavior the UI claims exists.
This includes:
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.
- 4d ago First seen · 214 lines · 0 tokens per session scan A 1a68f64853d3
OWNER_MODE_END_TO_END is an agent published in the GitHub repository HomenShum/NodeBenchAI (14 stars, last pushed 21d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,139 tokens. 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 agents, from other repositories
convex-expert
Expert Convex backend developer for queries, mutations, actions, schemas, auth, scheduling, components, and database operations. Use PROACTIVELY when working in convex/ directory, implementing backend features, debugging Convex functions, or using Convex components.
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.