Summary
PasswordPusher before version 2.9.2 exposes a brute-force weakness in the /p/:token/access endpoint, where passphrases can be guessed repeatedly without effective route-level defenses.
An attacker who knows a push token can submit around 120 guesses per minute, making short or dictionary-based passphrases realistically discoverable within hours or days.
Affected context
The vulnerability arises in deployments of PasswordPusher that:
- Rely on the
/p/:token/accessendpoint to gate access to password "pushes" using passphrases - Do not apply route-specific rate limiting or push-level lockout on failed attempts
- Allow an attacker to obtain or guess a valid push token (e.g., via link sharing or leakage)
In this context, the passphrase is treated as the main barrier to viewing a secret, but insufficient throttling means repeated online guessing remains feasible for weak passphrases.
Exploit preconditions
- Attack surface: Public or internal HTTP(S) interface exposing
POST /p/:token/access. - Required access: Knowledge of a valid push token; no prior authentication may be needed beyond that token.
- Trigger point: Automated or scripted POST requests submitting repeated passphrase guesses to the same push.
- Why exploitation is possible: The endpoint lacks granular rate limits and per-push lockout after consecutive failures, allowing sustained high-frequency guessing without meaningful penalties or alarms.
Why this happens
This issue is rooted in missing or insufficient brute-force defenses for a security-sensitive endpoint, matching CWE-307: Improper Restriction of Excessive Authentication Attempts.
The design treats the push token and passphrase as combined access controls, yet the application does not enforce per-resource throttling, step-up friction, or temporary lockout when multiple incorrect passphrases are submitted.
As a result, the system effectively permits unbounded, high-rate online guessing, eroding the effective entropy of user-chosen passphrases and making weak combinations practically discoverable.
Potential impact
- Unauthorized access to secrets or credentials stored in individual pushes
- Compromise of downstream systems if retrieved passwords are reused elsewhere
- Exposure of sensitive operational or personal data shared via PasswordPusher
- Erosion of trust in the platform's ability to safeguard time-limited or passphrase-protected secrets
General mitigation guidance
Effective mitigation typically focuses on strengthening online guessing resistance around sensitive endpoints and resources:
- Introduce route-specific rate limiting for
/p/:token/access, tuned to realistic human behavior rather than automated throughput. - Implement per-push lockout or backoff (e.g., exponential delays or temporary blocking) after a defined number of failed attempts.
- Add progressive friction such as CAPTCHAs or additional checks after repeated failures, especially on anonymous or unauthenticated access.
- Encourage or enforce high-entropy passphrases for protected pushes, reducing susceptibility to dictionary attacks.
- Apply least-privilege handling of secrets, ensuring that even if a push is compromised, blast radius is minimized (e.g., single-use credentials, short validity windows).
How teams detect issues like this
Organizations can look for this vulnerability and its abuse through a combination of review and monitoring:
- Static and configuration review: Check routing and middleware configurations to confirm that authentication or passphrase-protected endpoints have explicit rate-limiting and lockout logic.
- Code review signals: Identify security-sensitive access checks that rely solely on shared secrets or passphrases, but lack accompanying throttling, backoff, or anomaly detection.
- Log and behavior analysis: Monitor for repeated failed access attempts against the same push token, especially at machine-rate intervals, and flag or block such patterns.
- Automated tooling: Utilize application security scanners or custom linting rules that highlight authentication or secret-access flows without brute-force protections.

