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/walidboulanouar/define-doneWrote 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/commands/walidboulanouar/define-done/performance)<a href="https://agentmods.dev/commands/walidboulanouar/define-done/performance"><img src="https://agentmods.dev/badge/commands/walidboulanouar/define-done/performance/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/commands/walidboulanouar/define-done/performance"><img src="https://agentmods.dev/badge/commands/walidboulanouar/define-done/performance.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 | $0.00024 | $0.01170 |
| Opus 5.5 | $0.00010 | $0.00468 |
| Sonnet 5.5 | $0.00005 | $0.00234 |
| Haiku 4.5 | $0.00002 | $0.00117 |
Grade A, and why
performance 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 1mo 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 — 146 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/performance
The configuration ritual. Run this once when starting Define Done.
Most people use Define Done badly because they never defined what "done" means for THEM. Ambitious founders think it means "ship a feature." Burnt-out founders think it means "answered the urgent thing." Both are wrong.
This command asks you 7 questions, audits your answers, and writes your personal Define Done config to 008_Builds/define-done/config.md.
Workflow
Ask the questions one at a time. Do not batch them. After each answer, store it. Do not editorialize.
Q1 — Your role today
"In one sentence, what role do you play in your company today? Not your title. What is the work only you can do?"
(Why: outcomes must serve that role, not the role you used to have.)
Q2 — The shipped artifact
"Pick one thing you shipped last week that made you feel 'this was a real day's work.' What was it?"
(Why: that artifact is your unit. Future outcomes must match its size and shape.)
Q3 — The fake-work test
"Pick one thing you did last week that felt productive but didn't move anything. What was it?"
(Why: this is your trap pattern. We will write a rule against it.)
Q4 — Your hard stop
"What time do you want to close the laptop, every day, no exceptions? Pick a number."
(Why: the hard stop is the foundation. If you don't pick a time, the system collapses into "until I feel done.")
Q5 — Your weekly metric
"By Friday at 5pm, what single number tells you the week was a win?"
Examples: "3 client builds shipped" / "5 LinkedIn posts published" / "10 discovery calls booked" / "1 product release deployed"
(Why: this is your North Star. All weekly outcomes ladder to it.)
Q6 — Your "and" trap
"What word or phrase do you catch yourself adding to outcomes that always doubles the work? (e.g. 'and also', 'plus', 'while I'm at it')"
(Why: this is your personal banned phrase. We will reject any outcome containing it.)
Q7 — Your dopamine pattern
"When you finish a day feeling 'great', is it because you (a) shipped one important thing, (b) shipped many small things, (c) avoided a fire, or (d) other?"
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.
- 1mo ago First seen · 146 lines · 24 tokens per session scan A ac45894144fa
performance is a command published in the GitHub repository walidboulanouar/define-done (10 stars, last pushed 4mo ago), licensed MIT. It adds 24 tokens to every session and 1,170 once invoked, about $0.0001 per session on Opus 5.5. 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 commands, from other repositories
save-prompt
Generalize the current discussion into a reusable prompt and save it as a file.
review-resume
Comprehensive PM resume review against 10 best practices — structure, impact metrics, keywords, and actionable feedback.
orchestrate
Do this task as an orchestrator. The aim is the same result at a fraction of the cost: in a long session, every token in this conversation is re-read on each later request, while a subagent's reading is discarded when it returns. The plugin's subagents run on Haiku 5.5, about 40× cheaper per token than Opus 5.5.
generate-report
Generate a professional HTML report from assessment results in the conversation.
release
Generate changelog, bump version, and create git tag.
esp-teach
One-time project setup -- discover hardware, find datasheets, persist context to CLAUDE.md.