metrics-review
Skill0 starsChanged 4 months agoMITView on GitHub
Run a weekly metrics review ritual with a north star metric, input drivers, week-over-week deltas, and one experiment per week. Use when the user wants to review KPIs, set up startup metrics, pick a north star metric, or asks which numbers to track.
3 other projects use a version of this.
Adopted by
Most recently taken up by
- nisarth/claude-skills-storetheir own version, first seen Nov 8, 2026
- anthropics/knowledge-work-pluginstheir own version, first seen Oct 13, 2026
- robisson/build-like-amazon-agent-skillstheir own version, first seen Oct 5, 2026
Other versions3 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 3
- anthropics/knowledge-work-plugins★ 35ksame name
- robisson/build-like-amazon-agent-skills★ 16same name
- nisarth/claude-skills-store★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository10 other files it gives its agents.
- cold-outreachSkill
- competitor-scanSkill
- customer-discoverySkill
- hiring-kitSkill
- investor-updateSkill
- landing-copySkill
- launch-planSkill
- mvp-cutSkill
- pitch-deck-reviewSkill
- pricing-strategySkill
The file
We split this skill into the rules it gives an agent
3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 3 edited versions we read keep these.
2
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 21
Write definitions down precisely ("active = performed a core action, not logged in").
- Very specific: line 25
Ask for this week's actuals (never estimate or extrapolate a missing number.
- Distinctive: line 55
Never let a good top-line hide a bad input (growth masking collapsing retention is the classic startup killer).
“Common” is measured: kept by at least 6 in 10 of the 3 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Metrics Review
Five numbers reviewed every week beat forty numbers reviewed never. This skill sets up the metric tree once, then runs a repeatable weekly review against it.
Setup mode (first run)
Pick the north star: the one metric that best represents value delivered to customers. Weekly active teams, questions answered, orders processed. Revenue is an output, usually not the north star at early stage. Push back on vanity candidates: signups, page views, and total registered users measure marketing, not value.
Pick 3-5 input metrics the team can directly move, that plausibly drive the north star:
- Acquisition: new activated users (activated, not registered)
- Retention: week-4 retention or repeat-usage rate
- Revenue: MRR and net revenue retention once anyone pays
- Health: burn and runway, always present, never skipped
Write definitions down precisely ("active = performed a core action, not logged in"). Changing a definition later requires restating history.
Weekly review mode
Ask for this week's actuals (never estimate or extrapolate a missing number; mark it [missing] and flag that the tracking gap is itself a finding). Then produce the one-page snapshot:
WEEK OF [date]
NORTH STAR: X (+/-Y% WoW) [8-week trend: sparkline or list]
INPUTS
- metric: value (+/-% WoW) note if >15% move either way
RUNWAY: X months (burn $Y)
WHAT MOVED AND WHY
1-3 sentences per significant move. A number with no explanation
is homework, not insight.
THIS WEEK'S EXPERIMENT
One change being tested, the metric it should move, and the
threshold that counts as success.
DECISIONS NEEDED
Anything the numbers say the founder must act on.Rules
- Any input metric moving more than 15% week over week gets investigated before celebrating or panicking; check for tracking bugs first.
- One experiment per week. Zero means drifting; three means none will be readable.
- Ratios over totals as the company grows: retention rate beats total users.
- Runway under 6 months changes the review's tone: flag it explicitly every week it is true.
- Never let a good top-line hide a bad input (growth masking collapsing retention is the classic startup killer).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.