access-control-testing
Skill47 starsChanged 5 months agoMITView on GitHub
Use when testing whether users can access data or actions they shouldn't — IDOR/BOLA, horizontal and vertical privilege escalation, multi-tenant isolation, forced browsing, function-level authorization (BFLA), and mass assignment. The #1 cause of real-world breaches.
In this repository15 other files it gives its agents.
- active-pentestSkill
- api-security-testingSkill
- authentication-testingSkill
- business-logic-testingSkill
- client-side-exploitationSkill
- file-upload-and-ssrfSkill
- injection-testingSkill
- pentest-reportingSkill
- recon-and-osintSkill
- secrets-management-auditSkill
- security-code-auditSkill
- security-hardeningSkill
- security-testingSkill
- threat-modelingSkill
- vulnerability-chainingSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 4 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 9
Verify that every object and every action enforces "is *this* caller allowed?" server-side.
- Distinctive: line 14
Test the second one on every object and every action — not just the UI-visible ones.
- Distinctive: line 24
Use test accounts you created on an app you own/are authorized to test.
- Distinctive: line 34
If any cell succeeds, that's a finding.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Access Control Testing (FIND/EXPLOIT — OWASP A01)
Overview
Verify that every object and every action enforces "is this caller allowed?" server-side. Broken access control is the most common and highest-impact real-world flaw — and the easiest to miss because the happy path works perfectly.
Core principle: Authentication proves who you are; authorization proves what you may do. Test the second one on every object and every action — not just the UI-visible ones.
Applies when / Skip when
- Applies when: the app has authentication, multiple users, roles, or multi-tenant data — any resource or action that should be restricted. (Almost every non-trivial app.)
- Skip when: the app is fully public/anonymous with no per-user or privileged data → N/A.
- If N/A: report "access-control-testing: N/A — no auth/multi-user surface" and stop. Don't invent access boundaries that don't exist.
⚠️ Authorization
Use test accounts you created on an app you own/are authorized to test. Two accounts (A and B) plus one admin/one low-priv account is the cleanest setup.
The test matrix (run every cell)
| Read | Write/Update | Delete | Admin action | |
| As another user (horizontal) | A reads B's object? | A edits B's? | A deletes B's? | — |
| As lower privilege (vertical) | user hits admin read? | user hits admin write? | — | user invokes admin fn? |
| Unauthenticated | no token → data? | no token → mutate? | — | no token → admin? |
If any cell succeeds, that's a finding.
Method
- Capture a legitimate request as user A (note every id, uuid, slug, filename).
- Swap the object reference to B's resource, keep A's session → IDOR/BOLA.
- Swap the session (remove token / use lower-priv token) on privileged endpoints → BFLA / vertical.
- Force-browse hidden routes (
/admin,/api/internal) with a normal account. - Mass assignment — add fields the UI never sends (
role,isAdmin,userId,price). - Check every method —
GETmay be guarded butPUT/DELETE/PATCHforgotten.
See references/access-control-tests.md for concrete request recipes and edge cases.
Quick reference — bug classes
| Class | Test |
| IDOR / BOLA | Change an id/uuid/filename to another user's; do you get their data? |
| Horizontal privesc | Act on another same-level user's resources. |
| Vertical privesc | Low-priv account reaches admin function/data. |
| BFLA | Admin endpoint guarded in UI but callable directly by anyone. |
| Multi-tenant break | Tenant A reads tenant B (swap org_id/tenant_id). |
| Mass assignment | Inject privileged fields into the request body. |
| Forced browsing | Guess/enumerate unlinked privileged routes. |
Output
Findings in the register: which account hit which endpoint/object, the exact swapped request, what was exposed, and horizontal vs vertical. Severity is almost always High+.
Hand-off
Confirmed on a running app → reproduce via active-pentest; chain with others in
vulnerability-chaining; fix via security-hardening (deny-by-default, server-side checks).
Common mistakes
- Testing only as admin — you'll never see the bug. Always test as the lower privilege.
- Trusting that a hidden/unlinked endpoint is safe — enumerate and call it directly.
- Checking
GETonly — mutations are where damage happens. - Assuming UUIDs are safe — they leak (in responses, logs, referrers) and are still IDOR-able.
- Forgetting predictable patterns: sequential ids, base64'd ids, guessable filenames.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.