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 skills add RBraga01/builder-product --skill metric-definitiongit clone --depth 1 https://github.com/RBraga01/builder-productWrote 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/skills/rbraga01/builder-product/metric-definition)<a href="https://agentmods.dev/skills/rbraga01/builder-product/metric-definition"><img src="https://agentmods.dev/badge/skills/rbraga01/builder-product/metric-definition/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/skills/rbraga01/builder-product/metric-definition"><img src="https://agentmods.dev/badge/skills/rbraga01/builder-product/metric-definition.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 | $0.00053 | $0.01507 |
| Opus 5 | $0.00026 | $0.00754 |
| Sonnet 5 | $0.00011 | $0.00301 |
| Haiku 4.5 | $0.00005 | $0.00151 |
Grade A, and why
metric-definition 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 10d 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 — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Metric Definition
The Law
A METRIC DEFINED AFTER THE BUILD DESCRIBES WHAT WAS BUILT, NOT WHETHER IT SHOULD HAVE BEEN.
"We'll figure out what to measure from the data" means you cannot distinguish success from noise, or know whether to keep, kill, or iterate the feature.
North star + guardrail + diagnostic metrics with baselines, defined before code IS a measurement plan.
When to Use
Trigger before:
- Writing any implementation code for a new feature
- Committing to a PRD (see prd-quality-gate — metrics must be defined there too)
- Launching an A/B test (see ab-test-design — uses these metric definitions)
- Presenting a launch plan to stakeholders
When NOT to Use
- Infrastructure changes with no user-visible behaviour
- Bug fixes where "fixed" is the metric (the issue is the acceptance criterion)
- Time-boxed research spikes (the output is knowledge, not a metric)
The Three Metric Types
Every feature needs all three types. Missing any one creates a measurement blind spot.
1 — North Star Metric
The single metric that answers "did this feature achieve its purpose?"
Properties:
- One metric only — if you have two north stars, you have no north star
- Directly reflects user value (not a proxy for it)
- Moves when the feature is working, and only when the feature is working
- Has a defined target and timeframe (see prd-quality-gate)
Common failure:
"We'll track engagement" — engagement is not a north star; engagement of what action, by which users, compared to what baseline?
2 — Guardrail Metrics
Metrics that must NOT get worse when the north star improves.
Every feature optimisation can create a trade-off. Guardrail metrics make the trade-offs explicit and non-negotiable.
Required: at least two guardrails:
Guardrail 1: [Core product metric] must not drop by more than X%
Guardrail 2: [User trust / satisfaction metric] must not drop below Y
Example:
Feature: improve onboarding completion rate (north star) Guardrail 1: D7 retention must not drop by more than 2pp Guardrail 2: Support ticket volume must not increase by more than 10%
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.
- 10d ago First seen · 153 lines · 53 tokens per session scan A ecd7f596a688
metric-definition is a skill published in the GitHub repository RBraga01/builder-product (2 stars, last pushed 2mo ago), licensed MIT. It adds 53 tokens to every session and 1,507 once invoked, about $0.0003 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 skills, from other repositories
ai-engineering-toolkit
6 production-ready AI engineering workflows: prompt evaluation (8-dimension scoring), context budget planning, RAG pipeline design, agent security audit (65-point checklist), eval harness building, and product sense coaching.
promptfoo-provider-setup
Configure promptfoo providers or redteam targets for hosted models, live HTTP APIs, Python/JavaScript local scripts, agent SDKs, or multi-input systems. Use when connecting promptfoo to the system under test, mapping vars, auth env vars, request bodies, response transforms, or static-code-derived provider wrappers. Do…
avoid-ai-writing
Audit and rewrite content to remove AI writing patterns ("AI-isms"). Use this skill when asked to "remove AI-isms," "clean up AI writing," "edit writing for AI patterns," "audit writing for AI tells," or "make this sound less like AI." Supports a detect-only mode, an edit-in-place mode for files, an optional voice…
unlazy
Enforces completion discipline for substantial autonomous work by writing acceptance gates before execution, decomposing work with the Depth Tree, running approved checks, and re-verifying evidence before reporting. Use when an agent faces a long or multi-part task, work that has returned half-done, an exhaustive…
prompt-master
Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other…
avoid-ai-writing-router
Use when a request combines AI-writing audit, rewrite, file editing, voice preservation, false-positive interpretation, verification, or when the user invokes Avoid AI Writing without naming a mode.