Skip to navigation

Threat reportVulnerabilityTL-2026-0710

Instagram Web Password-Reset Logic Bug — Unredacted Email & Phone Disclosure (June 2026)

highPATCHED

A business-logic flaw in Instagram's web-based password-reset/account-recovery flow returned fully unredacted email addresses and phone numbers in place of the normally masked recovery options. Triggered by initiating a standard password reset for any username, it was demonstrated against high-profile accounts (e.g. 'zuck').

Key facts

IdentifiersTL-2026-0710
Also known asInstagram Password-Reset Recovery-Option Disclosure
SeverityHIGH
Affected productsMeta Instagram Web (password-reset / account-recovery flow)
Weaknesses (CWE)CWE-200, CWE-359, CWE-639, CWE-285
MITRE ATT&CK techniquesT1110 Brute Force, T1111 Multi-Factor Authentication Interception, T1213 Data from Information Repositories, T1566 Phishing, T1585 Establish Accounts, T1586 Compromise Accounts, T1589 Gather Victim Identity Information, T1595 Active Scanning, T1598 Phishing for Information, T1606 Forge Web Credentials
MotivationUNKNOWN
Target sectorssocial-media, consumer, media-and-entertainment, high-profile-individuals
Target regionsGlobal
Detection rules9 (1 shown free)
Sources6 cited sources

Detection coverage

As of 2026-06-07, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-0710 across Splunk SPL, Microsoft KQL and Sigma, covering 13 indicator(s) of compromise. The whole corpus is readable without an account. Every threat comes with one complete detection in Splunk SPL, Microsoft KQL and Sigma, free for everyone; a free account adds daily TL Creds to unlock the rest. Paid plans add a larger daily allowance, correlation and the MCP server. Threadlinqs MCP server · View plans.

Free detection: Instagram Account-Recovery Selector Enumeration — Exploitation Phase

Detects abuse of the Instagram web account-recovery flow (TL-2026-0710 masking-failure) where a single source initiates high-volume 'forgot password' POSTs to /accounts/account_recovery_send_ajax/ against many distinct usernames to harvest unredacted email/phone selectors. Fires on enumeration cadence (>25 distinct usernames / 5m per source IP).

  • Languages: Splunk SPL, Microsoft KQL, Sigma
  • ATT&CK: T1595.003, T1589.002, T1598
  • Data sources: proxy, webserver, waf
  • Severity: high
  • Confidence: high

Splunk SPL

index=proxy sourcetype="pan:traffic" OR sourcetype="access_combined" uri_path="/accounts/account_recovery_send_ajax/" http_method=POST
| bin _time span=5m
| stats dc(form_username) AS distinct_users count AS reset_attempts values(form_username) AS usernames by _time, src_ip
| where distinct_users > 25
| eval reason="High-volume recovery-option enumeration (selector harvesting)"
| table _time src_ip distinct_users reset_attempts usernames reason
| sort - distinct_users

Microsoft KQL

W3CIISLog
| where TimeGenerated > ago(24h)
| where csMethod == "POST" and csUriStem has "/accounts/account_recovery_send_ajax/"
| extend username = extract(@"username=([^&]+)", 1, csUriQuery)
| summarize distinct_users = dcount(username), reset_attempts = count(), usernames = make_set(username, 50) by cIP, bin(TimeGenerated, 5m)
| where distinct_users > 25
| project TimeGenerated, cIP, distinct_users, reset_attempts, usernames

Sigma

title: High-Volume Instagram Account-Recovery Selector Enumeration
id: 4a1f8d2e-7c63-4b19-9e4a-1d2f3c4b5a60
status: experimental
description: Detects enumeration-style abuse of the Instagram web account-recovery endpoint to harvest unredacted recovery selectors (TL-2026-0710).
references:
    - https://cybersecuritynews.com/instagram-password-reset-user-phone/
author: Threadlinqs Team
date: 2026/06/07
tags:
    - attack.reconnaissance
    - attack.t1595.003
    - attack.t1589.002
    - attack.t1598
logsource:
    category: webserver
detection:
    selection:
        cs-method: 'POST'
        cs-uri-stem|contains: '/accounts/account_recovery_send_ajax/'
    timeframe: 5m
    condition: selection | count(cs-username) by c-ip > 25
falsepositives:
    - Password-manager bulk-recovery tools or legitimate account-support batch operations
level: high

Known false positives:

  • Legitimate account-support batch recovery operations
  • Password-manager bulk health checks
  • Security scanners with authorized scope

Tuning notes: Tune the distinct_users > 25 threshold to your traffic baseline; lower for high-value brand-protection monitoring. Exclude known CDN/NAT egress IPs and authorized support tooling. Add geo/ASN enrichment on src_ip to prioritize bulletproof-hosting and datacenter ranges.

8 more detection rules (Splunk SPL, Microsoft KQL, Sigma) · unlock with TL Creds. Sign up free to unlock the detection rules or query them with the Threadlinqs MCP server (Purple)
  • Recovery-Flow Enumeration from Single Source — Exploitation Phase · unlock with TL Creds
  • Targeted Phishing Using Harvested Recovery Selectors — Post-Exploitation Phase · unlock with TL Creds
  • SIM-Swap Staging & SMS-2FA Interception Following Selector Harvest — C2/Lateral Phase · unlock with TL Creds
  • Account-Recovery Enumeration Burst — Exploitation Phase (Sigma primary) · unlock with TL Creds
  • Recovery Selector Harvesting to Local Staging — Post-Exploitation Phase (Sigma primary) · unlock with TL Creds
  • Phishing Lure Referencing Leaked Recovery Contacts — C2/Lateral Phase (Sigma primary) · unlock with TL Creds
  • Unredacted Recovery-Option Response Leakage — Exploitation Detection · unlock with TL Creds
  • Harvested-Selector Weaponization: Credential Stuffing & ATO — C2/Lateral Phase · unlock with TL Creds

Indicators of compromise

As of 2026-06-07, Threadlinqs Intelligence tracks 13 indicators of compromise for TL-2026-0710.

Sample indicators (1 of 13)

  • Domain: instagram[.]com

Defanged, TLP:CLEAR. Full set of 13 indicators: unlock with TL Creds.

13 indicators of compromise · unlock with TL Creds. Compare plans

Technical analysis

Meta deployed an emergency hotfix within hours, but PoC screenshots circulated widely. No CVE assigned.

On June 6, 2026 security researchers publicly disclosed a business-logic vulnerability in Instagram's web-based password-reset and account-recovery flow. The account-recovery screen is designed to present only partially redacted recovery options (e.g. an email rendered as 'm***@fb.com' or a phone shown as a masked suffix) so a requesting party can choose where a reset code should be sent without learning the full contact value. Due to a flaw in how the recovery options were serialized to the client, the flow returned the fully unredacted email address(es) and complete phone number(s) associated with the target account instead of the masked forms.

The defect was a pure server-side authorization/output-encoding logic bug, not a memory-safety or injection issue: an unauthenticated party simply initiated the standard 'forgot password' flow for an arbitrary username and read the recovery options returned by the account-recovery endpoint. No password reset needed to be completed and no code needed to be intercepted — the masked-recovery selection step itself leaked the data. Researchers demonstrated the issue against well-known accounts including Meta CEO Mark Zuckerberg ('zuck'), and proof-of-concept screenshots circulated on social media via accounts such as @vxunderground and @Scot0xo. Some account responses enumerated multiple email addresses tied to a single account.

Meta deployed an emergency hotfix within hours of disclosure and stated: 'We fixed an issue that allowed an external party to request password reset emails for some Instagram users. There was no breach of our systems.' No CVE was assigned. While the live window was short, even brief exposure of unredacted recovery data is operationally significant: an attacker who harvested email/phone pairs during the window obtains durable selectors for highly tailored phishing, SIM-swap attacks against SMS-based 2FA, credential-stuffing target lists, and targeted account-takeover (ATO). This incident follows a January 2026 wave of mass unsolicited Instagram password-reset emails and a parallel scraped-dataset ('17.5M records') circulating publicly, which together amplify the downstream phishing and doxxing risk of the leaked recovery selectors.

Because exploitation rides entirely on the legitimate Instagram recovery flow, defensive value lies in detecting abusive reset-initiation patterns (high-volume or enumeration-style 'forgot password' requests against many usernames), monitoring for the resulting targeted phishing using harvested unredacted contacts, and hardening account-recovery channels (authenticator-app 2FA over SMS, secured linked email).

MITRE ATT&CK techniques by tactic:

Credential Access

T1110 Brute Force; T1111 Multi-Factor Authentication Interception; T1606 Forge Web Credentials

Collection

T1213 Data from Information Repositories

Initial Access

T1566 Phishing

Resource Development

T1585 Establish Accounts; T1586 Compromise Accounts

Reconnaissance

T1589 Gather Victim Identity Information; T1595 Active Scanning; T1598 Phishing for Information

Remediation

Patches

  • Meta server-side hotfix (no client/app version dependency) deployed within hours of June 6, 2026 disclosure

Immediate actions

  • Meta hotfix already deployed server-side; no user action required to close the flaw
  • Treat any unsolicited Instagram password-reset email or SMS as suspicious and do not click links — initiate resets only from the official app/site
  • High-risk and high-profile accounts: rotate the linked recovery email/phone if they may have been harvested during the exposure window

Workarounds

  • None required post-fix; pre-fix there was no user-side workaround as the flaw was server-side in the recovery endpoint

Longer-term hardening

  • Migrate from SMS-based 2FA to an authenticator app or hardware key to blunt SIM-swap leverage
  • Secure the linked recovery email account with its own phishing-resistant 2FA
  • Periodically review Login Activity for unfamiliar sessions and devices

Affected and fixed versions

  • Meta — Instagram Web (password-reset / account-recovery flow)
    Vulnerable versions: web account-recovery flow as of 2026-06-06
    Fixed in: server-side hotfix deployed 2026-06-06

Timeline

  • Wave of ~1 million unsolicited Instagram password-reset emails reported; Meta acknowledges and fixes an issue allowing an external party to trigger reset emails, denies any breach.
  • Meta publicly states there was 'no breach' and that Instagram accounts are secure following the January reset-email wave.
  • Meta deploys an emergency server-side hotfix within hours of disclosure and issues a statement denying any system breach.
  • PoC screenshots circulate publicly (incl. via @vxunderground and @Scot0xo), demonstrating disclosure against high-profile accounts such as 'zuck'.
  • Researchers discover the web password-reset/account-recovery flow returns fully unredacted email addresses and phone numbers instead of masked recovery options.
  • Threadlinqs Intelligence ingests and publishes the threat for detection/IOC tracking; no CVE assigned.

Sources cited

Previous and next in vulnerability

Drafted by the Threadlinqs research pipeline from 6 sources; published ; last updated —.

Threadlinqs Intelligence — Real-Time Threat Detection Platform

// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level

Latest Threats