Skip to the rows

agentsclimarketplace

Index365 run audit

Skill index365usa/agent-skills/skills/index365-run-audit

Use when the user wants to run or re-run an index365 audit on a project. Triggers: "run an audit", "scan my site", "check my AI-readiness score", "re-run the audit", "run a marketing signal audit", "audit example.com again". Covers both audit types (AI-Readiness and Marketing Signal) and how to wait for the score. A run is a paid action, see below.From its SKILL.md

Install

npx -y skills add index365usa/agent-skills --skill index365-run-audit

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

3 things to look at

reads credentialsReads from 1 credential source: `INDEX365_API_KEY`.
1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
runs commandsInstructs the agent to run 5 commands, including `index365 login` and 4 more.

SKILL.md

2.5 KB, 552 tokens by cl100k_base, as published. Nobody here has run it

index365 run audit

Start an audit and (optionally) wait for the score.

Prerequisites

A projectId (from index365-add-project or index365 projects list).
An authenticated credential with runs:write (index365 doctor shows scopes). If authentication fails or the scope is missing, use index365-setup. If that skill is not installed, run index365 login (browser is the default choice; use --web or INDEX365_API_KEY without a terminal). After either recovery path, re-run index365 doctor before continuing. Do not work around missing authentication or scopes.

Pick the audit type

Run this	When the user cares about	scanMode
`index365 runs start`	how AI agents + AI search read the site (AI-Readiness)	`paid_ai_readiness`
`index365 marketing run`	whether demand can find, trust, act, and be measured (Marketing Signal)	`paid_marketing_signal`

If unsure, default to AI-Readiness; ask only if the user's intent is genuinely split.

Run and wait

# AI-Readiness, block until the score lands (~2–5 min; polls every 5s):
index365 runs start --project prj_xxx --wait --json

# Marketing Signal:
index365 marketing run --project prj_xxx --wait --json

--wait polls until the run reaches a terminal status (completed, failed, failed_auto_credit, refunded), then prints the final run with score and findingsTotal. Without --wait it returns immediately with a runId to poll via index365 runs get <runId>.
Record the runId and the baseline score, re-runs compare against it.

Paid action, retry safely

A run consumes credits. To make a retry safe (network blip, agent re-entry), pass a stable idempotency key so a repeat returns the same run instead of double-spending:

index365 runs start --project prj_xxx --idempotency-key "$(date +%F)-prj_xxx-airead" --wait --json

Next

Hand off to index365-read-report with the runId to read the score and findings, or index365-audit-and-fix to run, read, and fix in one loop.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Keep looking

Skills are one crate of 325,949. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.