experiments

An experiment-running guide for Phoenix, a tool for testing AI applications. It compares how prompts or pipelines perform across a dataset of test examples.

In plain words
What is it for?
Use it to run and compare dataset-backed experiments, inspect quality, latency, and cost, and judge whether an iteration helped.
Why use it?
It helps determine with evidence whether a change improved quality, speed, or cost instead of relying on a single result.

Skill for Claude CodeCodex

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 skills/arize-ai/phoenix/experiments
Any agent
npx skills add Arize-ai/phoenix --skill experiments
Clone the repo
git clone --depth 1 https://github.com/Arize-ai/phoenix

Made for: Claude Code, Codex.

Per session 164 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,460 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin unknown 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 $0.00164 $0.01460
Opus 5 $0.00082 $0.00730
Sonnet 5 $0.00033 $0.00292
Haiku 4.5 $0.00016 $0.00146

Measured 2d ago against content hash 5452705042ff, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

experiments 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 2d 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.

src/phoenix/server/agents/prompts/skills/experiments/SKILL.md · 93 lines

The source is not reproduced here

No licence file

A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.

Read it on GitHub

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. 2d ago First seen · 93 lines · 164 tokens per session scan A 5452705042ff

Subscribe to this mod's changes

experiments is a skill published in the GitHub repository Arize-ai/phoenix (11,275 stars, last pushed yesterday), with no licence file. It adds 164 tokens to every session and 1,460 once invoked, about $0.0008 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 2026-08-30.

Related

Other skills, from other repositories

trulens-running-evaluations

Execute TruLens evaluations and view results.

truera/trulens · 15 tokens

api-endpoints

Adding or changing API operations in @repo/operations. One source of truth (defineOperation + a Zod schema) becomes an HTTP endpoint, an OpenAPI operation, an MCP tool, TS + Python SDK methods, a latitude CLI command, and an in-process agent tool — descriptions and contracts must be written with all of these readers…

latitude-dev/latitude-llm · 78 tokens

fix-datadog-issues

Find, triage, and fix production errors captured by Datadog Error Tracking, then open a PR. Use when asked to look at "Datadog issues/incidents/errors", "find and fix bugs from Datadog", investigate the most-frequent or newest production errors, or work a specific Datadog Error Tracking issue.

latitude-dev/latitude-llm · 76 tokens

web-frontend

When to use: apps/web UI — routes, @repo/ui, TanStack Start server functions and collections, navigation (Link vs useNavigate), forms (useForm with createFormSubmitHandler + fieldErrorsAsStrings when Zod validation errors should appear on fields), Tailwind layout rules, design-system updates, and useEffect /…

latitude-dev/latitude-llm · 67 tokens

database-postgres

Drizzle schema, repositories, RLS, SqlClient wiring, Postgres migrations, psql / reset, or platform mappers (toDomain / toInsertRow).

latitude-dev/latitude-llm · 39 tokens

notifications

Multi-channel notifications. Adding a new notification kind, group, or channel; in-app + email delivery; per-user prefs; project-level gates; idempotency.

latitude-dev/latitude-llm · 35 tokens