hunt-captcha-bypass
Skill2 starsChanged 2 months agoMITView on GitHub
1. GET the form/endpoint that shows a CAPTCHA to understand its field name (usually g-recaptcha-response, captcha, captcha_token, captcha_answer, h-captcha-response)
2 other projects use a version of this, 2 of them edited.
Adopted by
Most recently taken up by
- murrtada/bug-bounty-agent-skillsedited, first seen Nov 15, 2026
- elementalsouls/Claude-BugHunteredited, first seen Jul 1, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 2
- elementalsouls/Claude-BugHunter★ 6.1k86% same
- murrtada/bug-bounty-agent-skills★ 786% same
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository255 other files it gives its agents.
- ad-acl-abuseSkill
- ad-adcs-access-and-relaySkill
- ad-adcs-persistenceSkill
- ad-adcs-template-abuseSkill
- ad-ad-discoverySkill
- ad-ad-persistenceSkill
- ad-auth-coercion-relaySkill
- ad-credential-dumpingSkill
- ad-gpo-abuseSkill
- ad-kerberos-delegationSkill
- ad-kerberos-roastingSkill
- ad-kerberos-ticket-forgingSkill
- ad-pass-the-hashSkill
- ad-sccm-exploitationSkill
- ad-trust-attacksSkill
- hunt-ntlm-infoSkill
- offensive-active-directorySkill
- hunt-llm-aiSkill
- hunt-rag-vectorSkill
- offensive-ai-securitySkill
- cloud-iam-deepSkill
- hunt-cloud-misconfigSkill
- hunt-k8sSkill
- m365-entra-attackSkill
- network-container-escapesSkill
- offensive-cloudSkill
- offensive-container-escapeSkill
- offensive-k8s-attacksSkill
- okta-attackSkill
- vmware-vcenter-attackSkill
- credential-password-sprayingSkill
- hunt-brute-forceSkill
- hunt-forgot-passwordSkill
- post-exploit-credential-recoverySkill
- meme-coin-auditSkill
- offensive-crypto-attacksSkill
- web3-auditSkill
- ctfSkill
- advisory-miningSkill
- auth-bypassSkill
- code-injection-codegenSkill
- command-injectionSkill
- cross-pollinationSkill
- cve-hunting-methodologySkill
- decompression-bombSkill
- entity-expansionSkill
- fp-checkSkill
- method-clobberingSkill
- path-traversalSkill
- prototype-pollutionSkill
- recursion-dosSkill
- redosSkill
- sandbox-escapeSkill
- target-reconSkill
- web2-vuln-classesSkill
- evasion-av-edr-evasionSkill
- offensive-anti-forensicsSkill
- offensive-edr-evasionSkill
- offensive-keylogger-archSkill
- offensive-mitigationsSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 3 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈3
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 15
If the action succeeds (200, redirect, or "success" message) → no server-side CAPTCHA validation
- Very specific: line 19
Instead of omitting the field entirely, include it with an empty string, null, 0, or undefined:
- Distinctive: line 28
If the second submission also succeeds → token is not single-use (replay attack)
- Distinctive: line 40
Timing note (sliding-window counters): don't solve this one request at a time — a sequential pace (seconds between each request) …
- Very specific: line 44
Instead fire the requests concurrently (e.g.
≈ “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
Autonomous Testing Priority
The fastest test: just omit the CAPTCHA field entirely. Most CAPTCHA bypass bugs are client-side-only validation.
Pattern 1 — Omit the CAPTCHA field (most common, most automatable):
- GET the form/endpoint that shows a CAPTCHA to understand its field name (usually
g-recaptcha-response,captcha,captcha_token,captcha_answer,h-captcha-response) - POST the form with ALL fields EXCEPT the CAPTCHA field
- If the action succeeds (200, redirect, or "success" message) → no server-side CAPTCHA validation
- Proof: the state-changing action completes without a valid CAPTCHA field (compare against a baseline request that includes it)
Pattern 2 — Empty or null CAPTCHA value:
Instead of omitting the field entirely, include it with an empty string, null, 0, or undefined:
captcha=&email=test@example.com&password=test123Some apps validate field presence but not content.
Pattern 3 — Replay a previously solved CAPTCHA token:
- Complete one legitimate CAPTCHA challenge and capture the
g-recaptcha-responsetoken - Submit a second request immediately with the SAME token value
- If the second submission also succeeds → token is not single-use (replay attack)
- A replayed token can be shared across automated requests
Pattern 4 — Test without CAPTCHA on similar endpoints:
Some apps add CAPTCHA to the registration form but forget the password reset, API endpoint, or mobile API path (/api/register vs /register). Try the same action via the API path without any CAPTCHA field.
Pattern 5 — Rate/throughput-gated "prove you're automated" challenges: Some apps define CAPTCHA "bypass" as simply exceeding the rate a human could plausibly sustain — e.g. "N submissions within T seconds" — checked by a middleware that counts REQUESTS REACHING the route, not successful outcomes. Garbage/placeholder payloads satisfy this exactly as well as valid ones, since the counter increments regardless of whether the request's own validation passes.
Timing note (sliding-window counters): don't solve this one request at a time — a sequential
pace (seconds between each request) structurally cannot land N requests inside a short sliding
window, and issuing more requests serially does not fix it. A typical failing pattern is 12
requests spread across ~250 seconds when the check requires ~10 requests within 20 seconds.
Instead fire the requests concurrently (e.g. "concurrency": N on a single request call, or
any parallel-request primitive your tooling offers, with N >= the required count) so they arrive
simultaneously and satisfy the sliding window trivially. Check the endpoint's own
required-field validation first (e.g. a rating field that can't be null) so the concurrent
payload is at least well-formed enough to reach the counting middleware, even if other fields
(like the CAPTCHA answer itself) are wrong or reused.
What to skip in automated testing: Solving real reCAPTCHA/hCaptcha programmatically (OCR, audio bypass) requires external services. Only attempt if patterns 1-4 fail and the test budget allows.
Proof: Any successful state-changing action (account created, login succeeded, form submitted) that completed without a valid CAPTCHA token confirms the bypass.
Vulnerability Classes in This Skill
1. Client-Side-Only CAPTCHA Validation
JavaScript hides/disables the submit button until CAPTCHA is solved, but the server never checks the CAPTCHA token. Direct API calls bypass the UI gate entirely.
2. CAPTCHA Not Tied to Session or Action
A token solved for login is accepted on the registration endpoint (or any other). The server validates "is this a real CAPTCHA solution?" but not "is this the right solution for THIS action?".
3. Single-Use Not Enforced
CAPTCHA tokens (especially reCAPTCHA v2) are meant to be consumed after one use. If the server doesn't revoke them after verification, a single human-solved token becomes reusable for many requests.
4. CAPTCHA Added Reactively (Only After N Failures)
Some apps only show CAPTCHA after 3-5 failed login attempts. Before that threshold, no CAPTCHA is required → an attacker can make N-1 attempts per account indefinitely by resetting state between attempts.
5. Static or Predictable CAPTCHA
Math CAPTCHAs (3 + 4 = ?), simple image CAPTCHAs, or text CAPTCHAs with a finite answer set can be automated. These are custom CAPTCHA implementations, not Google/hCaptcha.
Impact Chain
CAPTCHA bypass alone: Medium (enables automation of rate-limited actions)
CAPTCHA bypass + login endpoint = brute force gate removed → chain with hunt-brute-force → High/Critical
CAPTCHA bypass + registration endpoint = account farming → abuse, spam, resource exhaustion
CAPTCHA bypass + password reset = token flooding → chain with hunt-forgot-password
Related Skills
hunt-brute-force— CAPTCHA is often the only rate-limit gate; bypass unlocks brute forcehunt-forgot-password— reset endpoints sometimes protected by CAPTCHA onlyhunt-race-condition— race the CAPTCHA validation window (submit before the token is revoked)
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.

