Usage & Limits

Each nolapse plan enforces three types of limits: repository count, token count, and concurrent runner jobs. This page explains how each limit works, how usage is measured, and what happens when a limit is reached.

Repository Limit

A “repository” in nolapse is any directory initialised with nolapse init. Each call to nolapse init registers a repo slug with your organisation.

Plan	Repository limit
Free	10
Team	100
Enterprise	Unlimited

Running nolapse init on a repo that is already registered does not consume a new slot. The limit applies to distinct repos — counted by (org, repo_name) pair.

When you reach your repo limit, any attempt to initialise a new repo returns:

error: repository limit reached for your plan (10/10 repos used)
Upgrade to Team to track up to 100 repositories.

Existing repos are unaffected. They continue to accept nolapse run executions normally.

Token Limit

Plan	Tokens per organisation
Free	1
Team	10
Enterprise	Unlimited

The token limit applies to active tokens. Tokens in rotating status (within their grace period) count toward the limit. Revoked tokens do not.

When you reach the token limit, POST /v1/tokens returns 402 Payment Required:

{
  "error": "token limit reached for your plan (1/1 tokens used). Upgrade to create more tokens."
}

To free up a slot without upgrading, revoke an existing token with DELETE /v1/tokens/{id}.

Concurrent Runner Jobs

On the cloud deployment, nolapse processes nolapse run submissions through a job queue. The concurrency limit determines how many jobs execute simultaneously for your organisation.

Plan	Concurrent jobs
Free	5
Team	50
Enterprise	200

Jobs above the concurrency limit are queued, not dropped. They will begin executing as soon as a slot becomes available. In practice, queue time at typical usage is under 30 seconds on Free and negligible on Team and above.

For self-hosted deployments, there is no platform-enforced concurrency limit — it is bounded by your own infrastructure.

Execution History Retention

Each nolapse run produces an execution record accessible via GET /v1/executions. Records are retained for:

Plan	Retention period
Free	30 days
Team	90 days
Enterprise	1 year (longer by contract)

Records older than the retention window are deleted permanently. This affects the dashboard coverage history chart and the GET /v1/executions API. It does not affect the baseline file — baseline.md is stored in your git repository and is not subject to retention limits.

Checking Current Usage

You can check current token usage via:

curl -s \
  -H "Authorization: Bearer nlp_<your-token>" \
  https://api.nolapse.dev/v1/tokens | jq '.tokens | length'

Repository count will be available via a planned GET /v1/repos endpoint. For now, count the distinct repo values in your GET /v1/executions responses.

Upgrading to Remove Limits

If you are approaching a limit, upgrading your plan takes effect immediately. See Plans & Pricing for pricing details and how to upgrade.