Data practices

Privacy policy

Job Discovery Mail is a private personal-use application operated by its account owner. This policy describes the implemented collector and this public information website.

Effective date: October 12, 2026. Contact: daveallyn43.jobs@gmail.com.

Information accessed and stored

With the owner’s authorization, the app uses Gmail read-only access to verify the connected account and read message identifiers, sender and subject information, timestamps, and text/HTML message bodies. Collection covers the dedicated jobs mailbox within a time window, including Spam and Trash; it is not limited to selected senders or unread mail. The initial window is 30 days, followed by incremental scans with overlap.

The app stores derived job and source records: titles, employers, cleaned listing URLs, extracted context or short claims, message IDs, source labels, timestamps, processing status, unresolved leads and progress checkpoints. Public job descriptions and retrieval evidence may be added. Human review notes, if entered, are stored separately.

Full email bodies are processed in runtime memory and are not deliberately saved as raw email files, database bodies or workflow artifacts. Extracted claims can retain portions of message text. The collector does not fetch attachments, load email tracking images, or send, modify or delete messages.

Purpose and service providers

Information is used to discover, deduplicate and review job opportunities, preserve their provenance, and troubleshoot collection. Google supplies Gmail and OAuth authorization. GitHub Actions processes messages on hosted runners; GitHub Actions secrets hold the approved OAuth and database credentials. Derived records are stored in private Supabase PostgreSQL tables. Owner-operated local tools may retain authorization files and local development records.

Supported public publishers and job-board services receive requests for clean public article or job URLs. The collector does not send them full email bodies or intentionally follow recipient-specific tracking, unsubscribe or action links. These services can receive ordinary network request information.

During owner-directed review or troubleshooting, selected records or excerpts may also be processed by OpenAI assistant tools. This is separate from the scheduled collector, which uses deterministic parsing and does not call a language-model API or train models.

Job Discovery Mail does not sell Gmail data, use it for advertising, or use it to train generalized AI models. Google user data is used only for the stated job-discovery functions, consistent with the Google API Services User Data Policy, including its Limited Use requirements.

Retention and access controls

Derived records and review history remain until the owner deletes them; there is no automatic record-expiry policy. Reprocessing can retain earlier source claims. Public-source evidence artifacts currently have a seven-day GitHub retention setting. That limit does not apply to database records, local copies, secrets or workflow logs and summaries; provider retention and backup settings also apply.

The database is private, with restricted ingestion permissions and no anonymous job-table access. The collector cannot access or overwrite human feedback. These controls reduce exposure but do not guarantee absolute security.

Stopping access and deleting records

The owner can revoke the app’s access through their Google Account connections settings and stop the collection workflow. Revocation prevents future authorized Gmail reads but does not erase existing records. The owner must separately remove stored records, local copies and saved credentials through the relevant services. There is no automated deletion interface; provider backups may remain subject to their retention processes.

This website and changes

This website uses GitHub Pages hosting and the Porkbun-registered domain daveallynlab.com. Its files contain no private job records, forms, application analytics, advertising scripts or tracking cookies. Hosting and DNS providers may process ordinary connection information under their own policies. Material changes to these practices should be reflected here before the changed processing begins.