# Glossary > Plain definitions of the words used in the Virlo API docs, including three that mean different things on different pages: outlier, weighted_score, and run. Source: https://dev.virlo.ai/docs/glossary Markdown: https://dev.virlo.ai/docs/glossary.md Section: Start here ## About the Virlo API (applies to every page) - Base URL: `https://api.virlo.ai/v1`. Every request needs the header `Authorization: Bearer YOUR_API_KEY` (keys start with `virlo_tkn_`). - Responses are JSON inside a `data` field, except the `/v1/webhooks` endpoints, which return the object or array directly. Field names are snake_case. - Prices are in US dollars from a prepaid balance. 1 credit = $0.01. The `X-Cost` response header on each successful response is the exact charge. Errors are free. - Slow jobs return an ID. Check its status every 15 seconds (or whatever `retry_after_seconds` says) until `finalized` is `true`. - All docs pages: https://dev.virlo.ai/llms.txt. Every page in one file: https://dev.virlo.ai/llms-full.txt. MCP server for AI assistants: https://dev.virlo.ai/api/mcp/mcp. --- Virlo terms in plain words. [Three](https://dev.virlo.ai/docs/glossary#words-that-change-meaning) change meaning between pages. ## Research - **[Content Research Agent](https://dev.virlo.ai/docs/agents)**: collects short-form videos on your topic from TikTok, YouTube, and Instagram, then reports what's working. - **[One-time or recurring](https://dev.virlo.ai/docs/agents#create-agent)** (`is_recurring`): a one-time agent runs once, for $0.50 at creation. A recurring agent costs $0.50 per run. - **Cadence**: how often a recurring agent runs: `daily`, `weekly`, `monthly`, or a cron schedule up to once a day. Tracking uses `scrape_cadence`. - **[Data Intelligence](https://dev.virlo.ai/docs/intelligence)**: an AI breakdown of each video, like its hook, format, and tone. It adds $1.00 per agent run, or $0.25 per creator lookup. ## Lookups - **[Lookup](https://dev.virlo.ai/docs/satellite)**: a one-off deep dive on a creator, sound, hashtag, or video, for $0.50 to $2.50. Paths say `satellite`, the API's name for lookups. ## Tracking - **[Check](https://dev.virlo.ai/docs/tracking#how-it-works)** (also _cycle_ or _scrape_): one $0.25 round where Virlo fetches a tracked creator's or video's latest numbers and writes an AI report. - **[Snapshot](https://dev.virlo.ai/docs/tracking#get-creator-snapshots)**: the numbers saved at one check. An _audience snapshot_ is different: a $0.50 profile of a creator's commenters. ## Billing - **[Credit](https://dev.virlo.ai/docs/credits#response-headers)**: 1 credit = $0.01. Some headers and fields count in credits: `X-Credits-Used: 50` equals `X-Cost: 0.50`. ## Technical - **[Job ID](https://dev.virlo.ai/docs/async-data#per-resource)**: the ID you get right away when you start background work (a lookup's `job_id`, an agent's `id`), used to check on it. - **[Polling](https://dev.virlo.ai/docs/async-data#polling-pattern)**: checking every 15 seconds, or whatever `retry_after_seconds` says, until the job is done. It's free. - **[Finalized](https://dev.virlo.ai/docs/async-data#finalized)**: `finalized: true` means the results are ready, including follow-ups like an agent's AI report. `status: "completed"` alone isn't enough. - **[Webhook](https://dev.virlo.ai/docs/webhooks)**: a message Virlo sends to your web address when something happens, like an agent run finishing, so you don't have to poll. - **[Response envelope](https://dev.virlo.ai/docs/errors)**: successful responses wrap results in `{"data": ...}`. Errors, and calls under `/v1/webhooks`, aren't wrapped. ## Words that change meaning **Outlier**: far more views than expected, and "expected" differs by page. The _median_ is a creator's typical views. - [Content Research Agents](https://dev.virlo.ai/docs/agents#get-agent-outliers) and [hooks](https://dev.virlo.ai/docs/hooks): `outlier_ratio` is views ÷ followers. Hooks use one video's views, creator outliers a creator's average. - [Creator lookup](https://dev.virlo.ai/docs/satellite#start-creator-lookup): a video over `outlier_threshold` times the median (default `2`). - [Video outlier check](https://dev.virlo.ai/docs/satellite/video-outlier): `outlier_score` is views ÷ median. `3` or more is `viral`, `10` or more `mega_viral`. - [Tracking](https://dev.virlo.ai/docs/tracking#get-creator-signals) and [webhooks](https://dev.virlo.ai/docs/webhooks#tracking-payload): a video over 3 times the median (a breakout video). **`weighted_score`**: two different scores share this name, so never compare them. - On Content Research Agents and hooks, it's the **Virality Score**: how far views beat followers, with extra credit for big accounts. 18 or more is strong, 35 or more exceptional. - On lookups, tracking, and webhooks, it ranks outlier videos, giving big creators more credit. **For developers: the formulas** Virality Score: ln(views ÷ followers) × ln(followers). Lookups and tracking: ln(views ÷ median) × ln(median). **Run**: on [Content Research Agents](https://dev.virlo.ai/docs/agents#list-agent-runs), one round of collecting videos. On [lookups](https://dev.virlo.ai/docs/satellite#durable-runs), a saved result, free to re-read any time. Each paid lookup makes a new run. Its `run_id` is the `job_id`. --- More in Start here: - [Introduction](https://dev.virlo.ai/docs.md) - [Quickstart](https://dev.virlo.ai/docs/quickstart.md) - [Authentication](https://dev.virlo.ai/docs/authentication.md) - [Billing and pricing](https://dev.virlo.ai/docs/credits.md)