Hacktoberfest 2026: the issues maintainers tagged for October, open and beginner-friendly. Browse Hacktoberfest issues

Define zero-telemetry and strict-offline contract for dictation and future local translation

Closed
#65 1 comment 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
35/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
linux, shell
Domain
backend, security

Research direction

Start by tracing the bootstrap release and dependency downloads, faster-whisper model acquisition, and dictation runtime paths; the issue does not name specific files or tests. Verify which endpoints are contacted and what data upstream libraries may send, then define and implement strict offline behavior and audit telemetry and update defaults. Done means documenting the verified data flow, requiring explicit model acquisition, and demonstrating with egress blocked that cached dictation and future local translation run without steady-state requests.

Written by the indexing model from the issue text.

Description

enhancement privacy

Privacy contract / future translation design

HermesKey is local-first. A static sweep of current code found no HermesKey analytics or transcript-upload endpoint, but bootstrap downloads a release/dependencies and faster-whisper may fetch a model on first use. Customers should be able to distinguish explicit setup/model downloads from any runtime network access and choose a verified offline mode. Planned local-LLM translation must not silently introduce remote inference, telemetry, prompt logging or content upload.

Acceptance

  • Document a precise network/data-flow inventory: endpoints required for install/model download, when they are contacted, and what data could be sent by upstream libraries (verify rather than promising third-party behavior). State clearly that neither dictation nor future translation is uploaded by HermesKey.
  • Provide an opt-in strict offline configuration after a model is cached or supplied locally; fail clearly if a missing model requires download rather than silently going online. Audit upstream telemetry/update-check defaults and disable what can be disabled without breaking explicit downloads.
  • For any future translation engine: local model/process only by default, no remote translation API or transcript/prompt telemetry, no persistent content logs or implicit retention, explicit consent for model acquisition. Test with network egress blocked that preloaded dictation/translation still work and that no requests are made during steady-state operation.

This is a privacy requirement and verification task; this audit did not discover an existing HermesKey telemetry endpoint or claim that transcription currently uploads audio.

Dominant language
Shell
Stars
0
Forks
0
Avg merge
16m
Merged PRs (30d)
40

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from graygarbage/HermesKey

All issues in graygarbage/HermesKey

Similar issues

More Shell/Bash issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.