- Introduced `critic`, an independent adversarial reviewer for security and correctness.
- Added `fable-orchestrator` to manage task routing and verification.
- Implemented `gauntlet-critic` for fresh-context evaluation of gauntlet rounds.
- Created `planner` for generating executable implementation plans with dependencies.
- Developed `security-auditor` for application security reviews and audits.
- Established `system-steward` to improve agent prompts and skills based on verified failures.
- Added `dev-loop` skill for autonomous development loops over repositories.
- Implemented `gauntlet-loop` skill for iterative quality benchmarking against reference standards.
- Updated project settings to utilize the new orchestrator agent.
- Created documentation for `GAUNTLET.md`, `PROGRESS.md`, and `REFERENCE_BAR.md` to track project status and quality benchmarks.
- Added detailed prompting style guide to enhance understanding of prompt patterns and agentic loops.
A stored key carried no record of which provider it belonged to. Options
saves {provider, model} without the key whenever the field is blank (which
it always is after a save, since it shows the encrypted badge instead), so
switching provider left the previous provider's key attached to the new one.
Every call then failed with that provider's own "Invalid API Key" while the
UI still showed a key as configured.
- types.ts: new `keyProvider` storage field, added to CONFIG_STORAGE_KEYS
- background.ts: keyProviderMismatch() guards the chat, COPY_AS and
stored-key LIST_MODELS paths; absent keyProvider (pre-upgrade) is allowed
- Options.tsx: stamps keyProvider on every save; drops the encrypted badge
and requires a new key when the saved one belongs to another provider or
is rejected; save-time guard messages are now actually rendered (they were
gated on modelsStatus === 'error' and never drew, so Save looked dead)
- providers.ts: providerLabel(); settings hint appended to 401/403 only;
listModels reports keyRejected and labels errors with the display name
- Anthropic: send anthropic-dangerous-direct-browser-access on the chat path
Docs: CLAUDE.md version-bump rule corrected — wxt.config.ts reads
pkg.version, so package.json is the only place to edit.
typecheck clean, 58/58 tests, build clean (281.72 kB, manifest 1.0.2).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
- Added a new "Prompt Builder" tab in the Popup for generating AI prompts with customizable parameters.
- Introduced new state variables for managing prompt styles, personas, formats, and models.
- Enhanced the Options page to fetch and display models based on the provided API key.
- Updated the actions and types to include the new 'prompt' action and its associated parameters.
- Implemented migration logic for legacy plaintext API keys to encrypted storage.
- Updated the getSystemPrompt function to incorporate prompt parameters for better instruction generation.
- Added tests for the new functionality, including context menu entries and prompt generation logic.
- typecheck now also runs tsc over tsconfig.test.json (test files were
never type-checked before).
- Delete tests/unit/background.test.ts, which only asserted the vitest
chrome mock's own behavior — every remaining test imports production
code from src/lib.
- Baseline mock gains runtime.id, storage.session/remove, contextMenus,
and tabs.sendMessage.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
CONTEXT_MENU_ENTRIES maps every menu id to {action, style, parentId,
title}, replacing the split('-') id parsing that only worked because no
name contained a hyphen. Foreign menu ids are now ignored explicitly.
Parity pinned by tests enumerating all 30 previous ids. The 10-char
toolbar trigger threshold is now the named MIN_SELECTION_LENGTH.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
src/lib/providers.ts holds a per-provider spec (url, headers, body,
extractor, default model) and a single callProvider() covering both the
action-prompt and custom-prompt families (~250 lines -> one path).
getSystemPrompt, fetchWithTimeout, and listModels move along with it;
background.ts drops from 553 to 130 lines.
Behavior pinned by 20 parity tests written from the old functions
(request shapes, headers, defaults, error strings). One deliberate
change: max_tokens now scales with input length (1024-8192) instead of
a hard-coded 1024, so Expand no longer truncates long selections.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Encrypt (Options) and decrypt (background) now share one module; wire
format unchanged. Tests include fixtures proving values encrypted by the
old inline Options code still decrypt, plus tamper/wrong-key cases. The
module documents the honest threat model (key co-located with ciphertext
= obfuscation, not encryption).
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
- src/lib/types.ts: message contract (both ANALYZE_TEXT shapes preserved),
config/response types, storage-key constants.
- src/lib/messaging.ts: single safeStorageGet/safeStorageSet/safeSendMessage/
isExtensionValid implementation replacing the three divergent copies in
Options, Popup, and content. Content script keeps its refresh-toast
behavior via an onContextInvalidated callback.
- New '@lib' import alias (wxt force-overwrites '~' and '@' to srcDir, so
those cannot point at ./src); wired in wxt.config, tsconfig, vitest.
- tests/unit/messaging.test.ts: 14 unit tests over the wrappers.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
