> **Shapeless capabilities index**
> Fetch https://shapelessai.com/llms.txt to discover every page before exploring further.
> Every page also answers Markdown: append `.md` to its path, or send `Accept: text/markdown`.

# Read performance

*Verified 2026-10-07. Source: https://shapelessai.com/capabilities/analytics*

Every published post with its latest samples and the 90-day series, one post's accrual curve against the account's median, follower counts with the weekly delta, and a refresh that calls the platforms for stale samples.

The lift number a post carries is its performance against its channel's median, which is what an agent should compare drafts to before proposing the next one.

## Ask in plain words

- "which post did best this month"
- "how are followers trending on Instagram"
- "refresh the numbers and tell me what worked"
- "compare my last five LinkedIn posts against my median"
- "which platform is growing my followers fastest"

## From an agent host (MCP)

No MCP tool yet: call the REST routes below.

## From code (REST)

Bearer API key with the scope each call names ([auth](https://shapelessai.com/docs/auth), [OpenAPI](https://shapelessai.com/api/openapi.json)).

- `GET /api/metrics` (read scope): Published posts with their latest samples and the 90-day series.
- `GET /api/metrics/history` (read scope): One post's accrual curve against the account median (?post=).
- `GET /api/metrics/followers` (read scope): Follower counts and the weekly delta.
- `POST /api/metrics/refresh` (write scope): Refresh stale samples (calls the platforms).

## From a terminal (CLI)

No CLI command yet.

## Example

```bash
curl -s https://shapelessai.com/api/metrics -H "Authorization: Bearer $SHAPELESS_API_KEY"
```

## Where else

- [Do it in the studio](https://shapelessai.com/studio/performance)