Sanitized facts
Endpoint list, current limits, sanitized request summaries, latency/error snapshots, and shared dependencies.
The API Abuse-Budget Map is a tiny proof-backed listing format: one page that shows which endpoints can turn a normal-looking client into shared dependency damage, tenant spillover, support pain, or outage risk.
Endpoint list, current limits, sanitized request summaries, latency/error snapshots, and shared dependencies.
A risk table that ranks which endpoints can create visible damage and what evidence supports that claim.
No passwords, API keys, admin access, payment data, raw customer data, or destructive traffic generation.
| Column | Why It Matters |
|---|---|
| Endpoint | Names the surface being reviewed. |
| Operation type | Separates read, write, export, mutation, auth, and admin paths. |
| Auth scope | Shows whether anonymous, user, tenant, admin, or service auth is involved. |
| Current limit | Makes missing or weak enforcement visible. |
| Worst observed burst | Shows actual pressure, not theoretical risk. |
| Shared dependency touched | Identifies database, queue, cache, search, or external API coupling. |
| First saturation signal | Names what fails first: queue growth, latency, write amplification, or tenant spillover. |
| Blast radius | Shows whether one client can affect one user, one tenant, all tenants, or the platform. |
| Evidence path | Shows where the claim comes from: logs, config, dashboard, trace, or incident note. |
| Recommended default | Gives a safe starting limit or guardrail. |
| Exception owner | Prevents temporary exceptions from becoming permanent risk. |
The question is not "do we have rate limits?" It is "which endpoint lets one client spend everyone else's reliability budget?"
If yes, the next question is what sanitized input you would feel safe sharing.
If yes, the next question is what you would charge for a 24-hour version.
If no, the useful answer is the missing proof column that would make it trustworthy.