Skip to content

Conversation

@julwrites
Copy link
Owner

No description provided.

julwrites and others added 26 commits December 7, 2025 14:17
- Change search and ask result formatting to not use markdown links
- Disable /ask command for non-admins
- Prevent open-ended AI questions by requiring a bible reference
- Update tests to reflect the changes and mock API calls
- Implement task management via `scripts/tasks.py`.
- Create `docs/` structure (`architecture`, `features`, `tasks`).
- Update `README.md` with current status and local dev instructions.
- Update `AGENTS.md` to reflect maintenance mode.
- Create initial tasks for migration and cleanup.
Documentation Overhaul: Adopt Vibe Bootstrapper Standards
- Modified `GetBibleAsk` in `pkg/app/ask.go` to fall back to `ProcessNaturalLanguage` if the user is not an admin, instead of returning an error message.
- Modified `DumpUserList` and `Migrate` in `pkg/app/admin.go` to similarly fall back to `ProcessNaturalLanguage`.
- Added `"am": "Amos"` to `BibleBooks` map in `pkg/app/bible_reference.go` to restore support for the "Am" abbreviation and fix the regression in parsing logic.
- Verified that `TestParseBibleReference_Strict` passes with this change.
- Updated `TestGetBibleAsk/Non-admin_user` to expect search results (fallback behavior) instead of permission denied, aligning with current code logic.
- Used `SetAPIConfigOverride` and `ResetAPIConfigCache` to inject mock API configuration for deterministic testing.
- Updated `TestGetBibleAsk/Admin_user` to verify specific mock responses.
Fix TestGetBibleAsk failures using mock API config
Admin Silent Fallback and Amos Abbreviation Fix
- Refactored `SubmitQuery` in `pkg/app/api_client.go` to be a package-level variable for easier mocking.
- Added `MockSubmitQuery` helper in `pkg/app/test_utils_mock.go`.
- Updated `pkg/app/passage_test.go` to verify `Query.Verses` is populated correctly.
- Updated `pkg/app/search_test.go` to verify `Query.Words` is populated correctly.
- Updated `pkg/app/ask_test.go` to verify `Query.Prompt` and `Context.Verses` are populated correctly.
- Verified that natural language routing logic is covered by existing and updated tests.
…16093589716565048798

Verify API Request Arguments in Tests
- Updated `GetBibleSearch` in `pkg/app/search.go` to return HTML anchor tags for verses.
- Updated `pkg/app/search_test.go` to verify the new HTML format.
- Removed `secrets.yaml` generation in `.github/workflows/deployment.yml`.
- Removed `secrets.yaml` copy instruction in `Dockerfile`.
- Updated `GetBibleSearch` in `pkg/app/search.go` to return HTML anchor tags for verses.
- Updated `pkg/app/search_test.go` and `pkg/app/ask_test.go` to verify the new HTML format.
- Removed `secrets.yaml` generation in `.github/workflows/deployment.yml`.
- Removed `secrets.yaml` copy instruction in `Dockerfile`.
- Update `pkg/app/passage.go` to prevent extra newlines after spans when followed by `<br>`.
- Update `pkg/app/passage_test.go` with reproduction cases and updated expectations.
- Update `pkg/app/search.go` to return HTML links for search results.
- Update `pkg/app/search_test.go` and `pkg/app/ask_test.go` to match new search output.
- Remove `secrets.yaml` related code from `deployment.yml` and `Dockerfile`.
- Extracted `isNextSiblingBr` helper function in `pkg/app/passage.go`.
- Simplified `ParseNodesForPassage` to use the helper function instead of inline boolean state.
- Verified all tests in `pkg/app` pass.
…15737576654124430344

Fix search output format and remove dead secrets code
- Updated `go.mod` to use `BotPlatform@staging`.
- Created `pkg/utils/user.go` to define local `User` struct with state fields (`Action`, `Config`) and database tags, decoupling them from platform `UserData`.
- Added helper functions (`GetUserFromSession`, `SetUserAction`, `SetUserConfig`, `GetResourcePath`) to `pkg/utils/user.go` to manage user state and properties in `SessionData.Props`.
- Updated `pkg/utils/database.go` to use the local `User` struct for database operations.
- Refactored `pkg/bot/telegram.go` to map platform user to local user and store it in `env.Props`.
- Updated `pkg/app` logic to use helper functions instead of accessing removed fields (`env.User.Action`, `env.User.Config`, `env.ResourcePath`).
- Fixed `pkg/bot/sub.go` to correctly sync platform identity during subscription publishing.
- Updated all relevant tests to match the new architecture.
…ion-13807978232645792697

Refactor ScriptureBot for BotPlatform Decoupling
- Update BotPlatform dependency to master.
- Update `pkg/utils/user.go` `User` struct with `datastore:",noindex"` tags.
- Update `pkg/app/version.go` to use 3-column grid layout for version options.
- Update `pkg/app/database_integration_test.go` to gracefully skip if database credentials are missing.
…-16505421792742172411

Update ScriptureBot for BotPlatform compatibility
- Updated MIGRATION.md and README.md to include BIBLE_API_URL and BIBLE_API_KEY.
- Removed obsolete REVIEW_AND_PROPOSAL.md.
- Integrated Agent Harness structure:
    - Added scripts/tasks.py, scripts/memory.py, scripts/bootstrap.py, scripts/tasks.
    - Added .cursorrules and templates/maintenance_mode.md.
    - Added docs/tasks/GUIDE.md and docs/interop/tool_definitions.json.
    - Updated AGENTS.md and CLAUDE.md to match harness standards while preserving project instructions.
    - Created docs/memories/ directory.
…43751969655201

Update documentation and integrate Agent Harness
- Updated scripts/tasks.py, scripts/memory.py, scripts/bootstrap.py
- Updated templates/GUIDE.md and templates/maintenance_mode.md
- Updated AGENTS.md, .cursorrules, and .gitignore
- Added Next command to AGENTS.md
- Verified with task creation and deletion test
…281320230563

Update harness to latest version
…n-tools-17027230581161287134

Update documentation for new task management tools
@julwrites julwrites merged commit aaaa471 into master Dec 13, 2025
1 check passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants