metric-drift

metric-drift is a command for coding agents from mixpanel/ai-plugins. It costs 0 tokens per session (3,044 once invoked), scanned A, original, Apache-2.0.

A command that checks whether a metric's usual level has shifted over recent weeks, meaning the system may have entered a new trend or regime.

In plain words
What is it for?
Use it to compare a recent 30-day metric window with the preceding 30 days and judge whether the underlying trend has changed.
Why use it?
It distinguishes a lasting change in the baseline from a single unusual measurement, which needs a separate anomaly check.

Command

Part of the mixpanel plugin — 12 skills, 23 commands shipped together

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 commands/mixpanel/ai-plugins/metric-drift
Clone the repo
git clone --depth 1 https://github.com/mixpanel/ai-plugins

Or install mixpanel, the plugin that ships this one along with the rest of its 12 skills, 23 commands.

Wrote 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.

agentmods badge for metric-drift

README.md
[![agentmods](https://agentmods.dev/badge/commands/mixpanel/ai-plugins/metric-drift.svg)](https://agentmods.dev/commands/mixpanel/ai-plugins/metric-drift)
Your own site
<a href="https://agentmods.dev/commands/mixpanel/ai-plugins/metric-drift"><img src="https://agentmods.dev/badge/commands/mixpanel/ai-plugins/metric-drift.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,044 The whole file, excluding the scripts and references it only reads on demand.
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 $0.00000 $0.03044
Opus 5 $0.00000 $0.01522
Sonnet 5 $0.00000 $0.00609
Haiku 4.5 $0.00000 $0.00304

Measured 5d ago against content hash 504433c6c16b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

metric-drift 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 5d 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.

plugins/mixpanel/skills/monitor-metrics/commands/metric-drift.md · 258 lines

How it starts

The opening of the file, as written. The whole thing — 258 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Command: metric-drift

Detect trend-level drift in a single metric — whether the baseline itself has shifted over recent weeks. Produces a verdict on whether the metric is in a new regime. Does not test for point-in-time anomalies (run metric-anomaly for that).


Prerequisites

Before this command runs, Steps 0, 1, and 1.5 from references/execution.md must have completed — input validation, normalized metric series object, and project profile resolution. If any of those haven't happened, do them first.

If the user's input is a saved report but the metric is a funnel or retention report, see the "Special cases" section at the bottom.

Prerequisite — classify metric_type

Classify the metric per the metric_type table in SKILL.md and store metric_type on the series object before firing any queries.

Prerequisite — name the drift and baseline windows

The naming convention used throughout this command's output:

  • drift_window — the recent 30 days (most recent 30 days ending today).
  • baseline_window — the prior 30 days (30 days ending 30 days before today).

Both windows are computed from Q1-daily. The weekly test uses 8 vs 8 weeks — those windows are reported alongside but are secondary to the daily windows for headline purposes.


Phase 1 — Fetch series (2 queries, parallel)

Fire both queries simultaneously:

Query Window Granularity Comparison
Q1-daily Last 60 days day Last 30 days vs. prior 30 days
Q1-weekly Last 16 weeks week Last 8 weeks vs. prior 8 weeks

The 60-day daily view catches medium-term drift. The 16-week weekly view catches slow drift that the daily window would miss because daily noise drowns the signal. Running both is cheap and they answer different questions.

Use the query_template from the metric object; override only dateRange and unit (granularity). Do not re-apply filters — they're already baked in.

Read the full file on GitHub · 258 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. 5d ago First seen · 258 lines · 0 tokens per session scan A 504433c6c16b

Subscribe to this mod's changes

metric-drift is a command published in the GitHub repository mixpanel/ai-plugins (15 stars, last pushed yesterday), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 3,044 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.