{"id":51475361,"url":"https://github.com/asmuelle/inward","last_synced_at":"2026-07-06T20:30:33.334Z","repository":{"id":365985468,"uuid":"1266479361","full_name":"asmuelle/inward","owner":"asmuelle","description":"A voice-first journaling and reflection companion, verifiably airplane-mode functional — spoken thoughts are transcribed, reflected on, and stored only on the phone. Zero network in the journaling path.","archived":false,"fork":false,"pushed_at":"2026-07-01T07:43:48.000Z","size":3982,"stargazers_count":0,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-07-01T09:23:22.093Z","etag":null,"topics":["android","edge-ai","foundation-models","ios","journaling","on-device-inference","privacy","reflection","swift"],"latest_commit_sha":null,"homepage":"https://asmuelle.github.io/inward/","language":"Swift","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/asmuelle.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-06-11T16:50:34.000Z","updated_at":"2026-07-01T07:43:45.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/asmuelle/inward","commit_stats":null,"previous_names":["asmuelle/inward"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/asmuelle/inward","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/asmuelle%2Finward","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/asmuelle%2Finward/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/asmuelle%2Finward/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/asmuelle%2Finward/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/asmuelle","download_url":"https://codeload.github.com/asmuelle/inward/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/asmuelle%2Finward/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35205739,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-06T02:00:07.184Z","response_time":106,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["android","edge-ai","foundation-models","ios","journaling","on-device-inference","privacy","reflection","swift"],"created_at":"2026-07-06T20:30:32.292Z","updated_at":"2026-07-06T20:30:33.325Z","avatar_url":"https://github.com/asmuelle.png","language":"Swift","funding_links":[],"categories":[],"sub_categories":[],"readme":"# Inward\n\n[![CI](https://github.com/asmuelle/inward/actions/workflows/ci.yml/badge.svg)](https://github.com/asmuelle/inward/actions/workflows/ci.yml)\n\n\u003e A voice-first journaling and CBT-reframing companion that is verifiably airplane-mode functional — spoken thoughts are transcribed, reflected, and stored only on the phone.\n\n## Concept\n\nA voice-first journaling and CBT-reframing companion that is verifiably airplane-mode functional — spoken thoughts are transcribed, reflected, and stored only on the phone.\n\n\n##  Edge AI \n\n* On-device Whisper-class ASR for voice entries\n* AFM 3 Core with Dynamic Profiles running listener\n* CBT-reframe (cognitive distortions via @Generable structured output), and weekly-review modes\n* Spotlight local RAG retrieves past entries for longitudinal patterns\n* a LoRA adapter trained on reflective-questioning style as an uncopyable differentiator.\n* Android: Gemini Nano Summarization/Rewriting/Prompt APIs on flagships, Gemma 3n E2B via LiteRT-LM elsewhere. \n\n\n## Tech Stack\n\n* iOS (primary, iOS 26.4+): Swift/SwiftUI\n* SpeechAnalyzer + SpeechTranscriber for on-device ASR (whisper.cpp small as fallback for older devices)\n* FoundationModels AFM 3 Core via LanguageModelSession with @Generable structured output for distortion-tagging and reflection prompts, using the new context-size/token-count APIs to chunk under the 8K window with hierarchical entry summaries\n* Core Spotlight + NLContextualEmbedding (or a small Core ML embedding model) + sqlite-vec for local RAG over past entries; GRDB/SwiftData with SQLCipher and NSFileProtectionComplete; skip the custom LoRA adapter at launch (version-lock retraining tax) in favor of a prompt-engineered persona with few-shot exemplars; client-side-encrypted export to Files/iCloud Drive. Android (downscoped, flagships first): Kotlin/Jetpack Compose\n* ML Kit GenAI Summarization + Rewriting APIs on the AICore-supported device list, Prompt API only for short single-entry reflections (respect 4K-in/255-out limits); Gemma 3n E2B-it int4 via LiteRT-LM / MediaPipe LLM Inference as an opt-in download for non-AICore devices; on-device SpeechRecognizer or whisper.cpp for ASR\n* Room + SQLCipher\n* EmbeddingGemma-class embeddings via LiteRT + sqlite-vec for local retrieval. Both platforms: no analytics SDKs, no network calls in the journaling path (verifiable via iOS App Privacy Report)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fasmuelle%2Finward","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fasmuelle%2Finward","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fasmuelle%2Finward/lists"}