What therapists and clinicians should know about app-based recovery tools.

If you treat clients with compulsive porn use and have hesitated to recommend an app because of privacy concerns, against. is built specifically for the case you describe: local-only data, no accounts, no clinician portal.

Why most apps don’t fit clinical work

The main category of concern with consumer recovery apps is data handling. Most apps require account creation. Account creation means a server holds records associated with an identity. That server can be breached, subpoenaed, or acquired. The privacy policy, however thoughtfully written, is a promise, not a constraint.

For clinical work, this matters in several ways. A client may be involved in legal proceedings. A client may be in a relationship where data exposure would cause harm. A client may simply be deterred from honest logging if they believe the record could surface elsewhere. The app’s data handling has therapeutic consequences, not just technical ones.

A second category of concern is design philosophy. Many recovery apps are built around social features: accountability partners, community check-ins, public commitments. These features work for some people and in some contexts. For clients in therapy, the social layer may duplicate or interfere with the therapeutic relationship. The accountability structure is already present; an additional layer of app-mediated social accountability can be confusing or counterproductive.

A third concern is gamification. Streak counters and badges can undermine the kind of reflective engagement that supports clinical work. A client who experiences a relapse and fears the app’s visual feedback may avoid logging, precisely the moment when logging is most useful.

What against. does, technically

against. is a local-only tracker. All data is stored on the device, encrypted with AES-256 using a key generated on first run and stored in the device’s secure enclave. Nothing is transmitted to a server. There are no accounts and no user identifiers.

What the app records: relapse log entries (date, optional note), trigger annotations, mood or context tags, and streak count. The interface is minimal by design. A logging session takes under a minute.

What the app does not do: network sync, cloud backup, social features, account creation, analytics reporting, or any third-party data sharing. There is no SDK for advertising, attribution, or behavioral analytics.

The source code is available for review. The encryption implementation and schema are documented in the Privacy & Security section of the site.

How to use it as an adjunct

against. is most useful as a between-session tracking tool: a low-friction way for a client to record events close to when they occur, with enough structure to generate patterns that can be discussed in session.

Some clinicians find it useful to suggest the app as a way of collecting the kind of data that’s hard to reconstruct retrospectively: trigger context, time-of-day patterns, co-occurring mood states. The log isn’t intended to be shared with you; the client reads from it or summarizes. That distance can itself be clinically useful, since the client is the author of their own narrative.

If a client is working with a specific behavioral protocol (a counting method, a urge-surfing practice, a structured journaling model), the app’s notes field is flexible enough to support most annotation formats. It doesn’t impose a framework.

One thing to set expectations about: the app doesn’t send you data. There’s no integration, no dashboard, no notification. It’s a tool your client uses independently. The value to you comes through what they bring to the session.

What we don’t claim

against. is not a treatment. It’s not a substitute for therapy. It is not evidence-based in the sense that a specific therapeutic modality is evidence-based. It’s a tracking tool, and the evidence base is for self-monitoring in general, which is well-established but generic.

The app doesn’t diagnose. It doesn’t use clinical language about disorder or compulsion. That framing belongs in a clinical context, not an app. It uses the language of patterns and choices.

We don’t claim the app is appropriate for every client. For someone with significant shame or perfectionism, a streak-visible interface might not be the right fit. For someone in acute crisis, a tracking app is not the intervention. Clinical judgment about whether and when to suggest it belongs with you.

We also don’t claim to be the last word on private app design. The threat model we defend against is realistic but not unlimited. We document the tradeoffs honestly (see offline-first matters). A client with specific threat conditions (active legal exposure, high-control relationship, state-level surveillance concerns) may need more specialized guidance.

Further reading

• How encryption works in against.
• Why offline-first matters for sensitive data
• What the science says about habit-tracking and recovery

See also: For clinicians.