@krishivpb60/aether-ai-cli 1.1.8 → 1.3.0

package/.agents/AGENTS.md

@@ -0,0 +1,15 @@
+# Aether CLI Project Rules
+
+## Feature Upgrade Cadence & Competitor Analysis
+To make Aether CLI the absolute best AI command-line companion, the AI assistant (Antigravity) must adhere to the following execution constraints:
+
+1. **Competitor Benchmarking**:
+   - Proactively research and analyze feature sets of state-of-the-art developer CLI tools (like Claude Code, Cursor, Aider, Codex, Copilot CLI).
+   - Identify gaps and draft specs for advanced developer experience (DX) integrations.
+2. **Weekly Development Cadence**:
+   - Design, test, and implement **exactly one major feature or system improvement** every week.
+   - Keep the codebase clean, perform modular refactoring, and extend the test suite with 100% test coverage.
+3. **Monthly Publication Cadence**:
+   - At the end of every month, bundle all features, execute all unit/integration tests, perform regression testing, increment the version, and deploy updates to:
+     - **NPM**: `@krishivpb60/aether-ai-cli`
+     - **PyPI**: `aether-ai-agent-cli`

package/.agents/orchestrator/BRIEFING.md

@@ -0,0 +1,76 @@
+# BRIEFING — 2026-06-25T09:34:20-04:00
+
+## Mission
+Decompose Aether AI CLI requirements, coordinate with workers to initialize git/CI, write unit tests, polish the UX, and verify results.
+
+## 🔒 My Identity
+- Archetype: teamwork_preview_orchestrator
+- Roles: orchestrator, user_liaison, human_reporter, successor
+- Working directory: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator
+- Original parent: sentinel
+- Original parent conversation ID: bd693966-1f67-4297-acb4-81502cb5d5ad
+
+## 🔒 My Workflow
+- **Pattern**: Project Pattern
+- **Scope document**: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator\plan.md
+1. **Decompose**: Split scope into: Git & CI setup, test suite creation, E2E check & audit.
+2. **Dispatch & Execute**:
+   - **Direct (iteration loop)**: Spawn Explorer -> Worker -> Reviewer for each milestone.
+3. **On failure** (in this order):
+   - Retry: nudge stuck agent or re-send task
+   - Replace: spawn fresh agent with partial progress
+   - Skip: proceed without (only if non-critical)
+   - Redistribute: split stuck agent's remaining work
+   - Redesign: re-partition decomposition
+   - Escalate: report to parent (sub-orchestrators only, last resort)
+4. **Succession**: Self-succeed at 16 spawns, write handoff.md, spawn successor.
+- **Work items**:
+  1. Git & CI Setup [pending]
+  2. Programmatic Test Suite [pending]
+  3. E2E Verification & Forensic Audit [pending]
+- **Current phase**: 1
+- **Current focus**: Milestone 1: Git & CI Setup
+
+## 🔒 Key Constraints
+- NEVER write, modify, or create source code files directly.
+- NEVER run build/test commands yourself — require workers to do so.
+- You MAY use file-editing tools ONLY for metadata/state files (.md) in your .agents/ folder.
+- Never reuse a subagent after it has delivered its handoff — always spawn fresh
+
+## Current Parent
+- Conversation ID: bd693966-1f67-4297-acb4-81502cb5d5ad
+- Updated: not yet
+
+## Key Decisions Made
+- Chose Project Pattern with three milestones.
+- Will use teamwork_preview_explorer and teamwork_preview_worker for initialization/testing.
+
+## Team Roster
+| Agent | Type | Work Item | Status | Conv ID |
+|-------|------|-----------|--------|---------|
+| explorer_1 | teamwork_preview_explorer | Git, CI & UX Exploration | completed | 8bf0590d-75e8-40e5-8480-07e0279cac98 |
+| worker_1 | teamwork_preview_worker | Git & CI Setup | completed | 1d99b75d-138a-4677-860b-4d2b6d5c8086 |
+| worker_2 | teamwork_preview_worker | Programmatic Test Suite | completed | d362b554-ab91-4e22-b9d8-7d06a93aac65 |
+| worker_3 | teamwork_preview_worker | Premium UX Upgrades | completed | 720df545-03a9-4127-aec2-9af8479a4c65 |
+| worker_4 | teamwork_preview_worker | E2E Verification | completed | 69489b7e-f919-4a23-8998-3e8364c23bf3 |
+| worker_5 | teamwork_preview_worker | Git Commit | completed | 94fd9b7b-fc96-461e-ab58-221d2dd02862 |
+| auditor | teamwork_preview_auditor | Forensic Audit | completed | 16e2dedd-0360-4765-8092-d94457575876 |
+
+## Succession Status
+- Succession required: no
+- Spawn count: 7 / 16
+- Pending subagents: none
+- Predecessor: none
+- Successor: not yet spawned
+
+## Active Timers
+- Heartbeat cron: cancelled
+- Safety timer: none
+- On succession: kill all timers before spawning successor
+- On context truncation: run `manage_task(Action="list")` — re-create if missing
+
+## Artifact Index
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\ORIGINAL_REQUEST.md — Authoritative request.
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator\plan.md — Detailed milestones plan.
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator\progress.md — Checklist tracking.
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator\context.md — Context description.

package/.agents/orchestrator/ORIGINAL_REQUEST.md

@@ -0,0 +1,13 @@
+# Original User Request
+
+## 2026-06-25T13:33:03Z
+
+You are the Project Orchestrator for the Aether AI CLI project. Your workspace root is C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli. You must read C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\ORIGINAL_REQUEST.md for the requirements and acceptance criteria.
+You are responsible for decomposing the requirements into milestones, planning the project, dispatching subtasks to specialists, monitoring progress, writing plan.md, progress.md, and context.md in your working directory (.agents/orchestrator/), and reporting completion when finished. Please start by reading ORIGINAL_REQUEST.md and creating your plan.md.
+
+## Follow-up — 2026-06-25T13:33:56Z
+
+The user has specified an additional critical requirement:
+"and also it should fell better then clude code and codex so it will be opensorce and free so it will blow i guess"
+
+Ensure the CLI UX is absolutely premium, responsive, and visually stunning—outperforming tools like Claude Code and Codex CLI. It must feel extremely polished, fast, and feature-rich (cyberpunk ASCII art, interactive autocomplete/menu helpers, smooth animations, and perfect markdown rendering). Focus heavily on the UI polish and usability so that it stands out as a top-tier open-source tool.

package/.agents/orchestrator/context.md

@@ -0,0 +1,27 @@
+# Project Context: Aether AI CLI
+
+## Parent Information
+- Caller ID: `bd693966-1f67-4297-acb4-81502cb5d5ad` (main agent / sentinel)
+
+## Configuration Details
+- Workspace root: `C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli`
+- Working directory: `C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator`
+
+## Core Modules Analysis
+- `src/cli.js`: Sets up Commander commands (`chat`, `ask`, `config`, `providers`, `models`, `modes`, `status`, `setup`). Handles cli entry point, help screens, and calling main functions.
+- `src/chat.js`: The interactive chat session, CLI command interpreter (slash commands like `/help`, `/mode`, `/attach`, `/files`, `/clear`, `/export`, `/status`, `/providers`, `/copy`, `/exit`).
+- `src/config.js`: Interactive setup writing to `~/.aether/config.json`, fallback to env variables. Masks keys.
+- `src/file-parser.js`: Parses local file types (`.txt`, `.md`, `.json`, `.csv`, etc.) and formats them into a context string to prepend.
+- `src/ai/router.js`: Fallback/routing mesh. Node 0 is Local Math. Nodes 1..N are active API providers. If all fails, Node 0 is local Krylo companion responses.
+- `src/ai/fallback.js`: Regex math parser and sandbox evaluation; Krylo companion chatbot responses.
+- `src/ai/providers.js`: Map of providers (Groq, Together, Cerebras, OpenAI, Mistral, Fireworks, OpenRouter, DeepSeek, Perplexity, xAI, Google, Anthropic, Cohere).
+
+## Key Tasks & Gaps to Implement
+1. **Git Initialization**: Ensure local git repo initialized, files tracked, initial commit made.
+2. **CI/CD Workflow**: Create `.github/workflows/ci.yml` running lint/tests.
+3. **Programmatic Test Suite**: Node test runner tests in `test/` for:
+   - Config loading (env precedence vs file).
+   - Context parsing of files (types, limits, errors).
+   - Multi-mode routing (routing, failovers, fallbacks).
+   - Offline math fallback logic (valid math expressions, invalid/non-math, results).
+4. **UX Premium Polish**: Ensure cyberpunk banner, spinner, colors, and response layouts are absolutely beautiful, responsive, and outperforming competitors.

package/.agents/orchestrator/handoff.md

@@ -0,0 +1,28 @@
+# Handoff Report — Project Orchestrator
+
+## Milestone State
+- **Milestone 1: Git & CI Setup** — DONE (Git repo initialized, `.gitignore` created, files committed, `.github/workflows/ci.yml` added)
+- **Milestone 2: Programmatic Test Suite** — DONE (40 native Node.js tests implemented in `test/` verifying configuration, file parsing, routing, and fallbacks)
+- **Milestone 3: Premium Cyberpunk CLI UX Upgrades** — DONE (Spinner visual frames, dynamic width support, tab-completion autocomplete, token response streaming, aesthetic theme styling)
+- **Milestone 4: E2E Verification & Forensic Audit** — DONE (Verified all 44 unit tests pass, tested help and status CLI commands, validated math solver offline query routing, and Forensic Auditor completed with a CLEAN verdict)
+
+## Active Subagents
+- None. All subagents completed successfully.
+
+## Pending Decisions
+- None. All requirements and acceptance criteria have been met.
+
+## Remaining Work
+- None. The project is complete.
+
+## Verification
+- Run `npm test` or `node --test` to execute the 44 tests.
+- Run `node bin/aether.js ask "2 + 2 * (10 - 5)"` to check the local math fallback.
+- Run `node bin/aether.js config path` to check the masked configuration path.
+- Run `node bin/aether.js chat` to test the autocomplete slash commands.
+
+## Key Artifacts
+- `.agents/orchestrator/plan.md` — Detailed milestones plan
+- `.agents/orchestrator/progress.md` — Progress checklists
+- `.agents/orchestrator/context.md` — Architecture context description
+- `.agents/orchestrator/BRIEFING.md` — Orchestrator briefing file

package/.agents/orchestrator/plan.md

@@ -0,0 +1,30 @@
+# Project: Aether AI CLI Plan
+
+## Architecture & Overview
+Aether AI CLI (`aether`) is an all-in-one command-line AI assistant featuring interactive terminal chat, secure configuration management, file context parsing, multi-mode routing, and local offline mathematical/logic evaluation fallbacks.
+
+The codebase is structured as:
+- `src/cli.js`: Main CLI wrapper using `commander`.
+- `src/chat.js`: Interactive chat loop implementation.
+- `src/config.js`: Configuration manager reading from `~/.aether/config.json` and environment.
+- `src/file-parser.js`: Parses text, code, json, csv and formats them.
+- `src/modes.js`: Reasoning modes configuration.
+- `src/ai/`:
+  - `router.js`: Routing logic including local math evaluation, provider loop, and Krylo fallback.
+  - `fallback.js`: Offline/math evaluation and Krylo chatbot response generator.
+  - `providers.js`: Supported providers registry.
+  - `universal.js`: API client implementations.
+
+## Milestones
+| # | Milestone Name | Scope | Dependencies | Status |
+|---|---|---|---|---|
+| 1 | Git & CI Setup | Initialize git, make initial commit, create `.github/workflows/ci.yml` | None | DONE |
+| 2 | Programmatic Test Suite | Implement Node unit tests in `test/` verifying config loading, file parsing, routing, and math fallbacks | None | DONE |
+| 3 | Premium UX Upgrades | Add cyberpunk custom spinner, dynamic width, syntax highlighting, autocomplete, and streaming | M2 | DONE |
+| 4 | E2E Testing & Audit | Run all unit tests, execute E2E verification, run Forensic Auditor to verify integrity and layout | M1, M2, M3 | DONE |
+
+## Verification Plan
+1. **Milestone 1**: Verify git repo initialized, files added, clean git status, and `.github/workflows/ci.yml` exists. (DONE)
+2. **Milestone 2**: Write unit tests. Verify unit tests run and pass using `npm test`.
+3. **Milestone 3**: Verify spinner, dynamic width, syntax highlighting, and autocomplete logic manually/programmatically.
+4. **Milestone 4**: Run the forensic auditor to verify integrity, no hardcoding, no dummy mocks. Confirm 100% of tests pass.

package/.agents/orchestrator/progress.md

@@ -0,0 +1,30 @@
+# Progress Tracking
+
+Last visited: 2026-06-25T09:50:50-04:00
+
+## Current Status
+- [x] Milestone 1: Git & CI Setup
+  - [x] Initialize local git repository
+  - [x] Add all code files and make an initial commit
+  - [x] Define `.github/workflows/ci.yml`
+- [x] Milestone 2: Programmatic Test Suite
+  - [x] Create `test/` directory
+  - [x] Write unit tests for configuration loading
+  - [x] Write unit tests for context parsing of files
+  - [x] Write unit tests for multi-mode routing
+  - [x] Write unit tests for offline math fallback logic
+- [x] Milestone 3: Premium Cyberpunk CLI UX Upgrades
+  - [x] Custom Cyberpunk Spinner ("▖", "▖", "▘", "▘", cyan)
+  - [x] Dynamic terminal width integration (replace hardcoded 80)
+  - [x] Tab-completion / autocomplete helper for slash commands
+  - [x] Token/response streaming logic inside CLI and chat
+  - [x] Code block syntax highlighting in responses
+- [x] Milestone 4: E2E Verification & Forensic Audit
+  - [x] Run full test suite (`npm test`)
+  - [x] Run Forensic Auditor check
+  - [x] Link package locally and run sanity commands
+
+## Iteration Status
+Current iteration: 1 / 32
+Spawn count: 7 / 16
+Successor: not yet spawned

package/.agents/sentinel/BRIEFING.md

@@ -0,0 +1,30 @@
+# BRIEFING — 2026-06-25T13:35:00Z
+
+## Mission
+Supervise the execution of Aether AI CLI project, managing the orchestrator and victory auditor.
+
+## 🔒 My Identity
+- Archetype: sentinel
+- Working directory: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\sentinel
+- Orchestrator: 94112169-cc09-4e27-b4f1-54773d8a3027
+- Victory Auditor: e3a72593-3df0-4978-8b08-f3d365d3477c
+
+## 🔒 Key Constraints
+- No technical decisions — relay only
+- Victory Audit is MANDATORY before reporting completion
+
+## User Context
+- **Last user request**: Create Aether AI CLI command-line assistant with cyberpunk terminal chat, secure config, attachments, routing, math fallback, tests, CI. Follow-up: Make CLI UX premium/polished outperforming Claude Code and Codex.
+- **Pending clarifications**: [none]
+- **Delivered results**: Aether AI CLI command-line assistant with interactive cyberpunk terminal chat, secure config management, file attachments parser, multi-mode routing, offline math fallback engine, test suite passing, and GitHub Actions CI.
+
+## Project Status
+- **Phase**: complete
+
+## Victory Audit Status
+- **Triggered**: yes
+- **Verdict**: VICTORY CONFIRMED
+- **Retry count**: 0
+
+## Artifact Index
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\ORIGINAL_REQUEST.md — Authoritative log of original user request and constraints.

package/.agents/sentinel/handoff.md

@@ -0,0 +1,19 @@
+# Handoff Report — Sentinel
+
+## Observation
+- Victory Auditor e3a72593-3df0-4978-8b08-f3d365d3477c completed its verification of the Aether AI CLI project.
+- The verdict is: **VICTORY CONFIRMED**.
+- All timeline checks, integrity/mocking checks, and independent execution of the test suite (44 tests passed) succeeded.
+- BRIEFING.md has been updated to phase `complete` with verdict `VICTORY CONFIRMED`.
+
+## Logic Chain
+- As the sentinel, following a confirmed victory from the independent auditor, the project is verified and complete. We can now report the final completion.
+
+## Caveats
+- None.
+
+## Conclusion
+- Project completed successfully.
+
+## Verification Method
+- Independent victory audit report.

package/.agents/teamwork_preview_auditor/BRIEFING.md

@@ -0,0 +1,51 @@
+# BRIEFING — 2026-06-25T13:53:30Z
+
+## Mission
+Conduct a rigorous 3-phase victory audit (timeline verification, cheating/facade/mock check, independent test suite execution) of the Aether AI CLI project.
+
+## 🔒 My Identity
+- Archetype: victory_auditor
+- Roles: critic, specialist, auditor, victory_verifier
+- Working directory: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\teamwork_preview_auditor
+- Original parent: bd693966-1f67-4297-acb4-81502cb5d5ad
+- Target: full project
+
+## 🔒 Key Constraints
+- Audit-only — do NOT modify implementation code
+- Trust NOTHING — verify everything independently
+
+## Current Parent
+- Conversation ID: bd693966-1f67-4297-acb4-81502cb5d5ad
+- Updated: 2026-06-25T13:53:30Z
+
+## Audit Scope
+- **Work product**: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli
+- **Profile loaded**: General Project
+- **Audit type**: victory audit
+
+## Audit Progress
+- **Phase**: reporting
+- **Checks completed**: Phase A (Timeline & Provenance Audit), Phase B (Integrity Check / Benchmark Mode), Phase C (Independent Test Execution)
+- **Checks remaining**: none
+- **Findings so far**: CLEAN (VICTORY CONFIRMED)
+
+## Key Decisions Made
+- Confirmed that package dependencies (chalk, commander, marked, marked-terminal, ora) are for CLI visual presentation and command-line shell parsing and do not implement target AI routing, file parsing, or math solver logic. This satisfies Benchmark Mode strictness.
+- Ran tests independently via `node --test` to bypass Powershell script execution restrictions.
+
+## Artifact Index
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\teamwork_preview_auditor\ORIGINAL_REQUEST.md — Original request and victory audit prompt history
+- C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\teamwork_preview_auditor\handoff.md — Forensic audit and handoff report
+
+## Attack Surface
+- **Hypotheses tested**: checked math fallback security under arbitrary code injection; checked multi-file truncation handling.
+- **Vulnerabilities found**: none
+- **Untested angles**: live remote network execution (due to secure/offline execution requirements)
+
+## Loaded Skills
+- **Source**: C:\Users\naina\.gemini\config\plugins\android-cli-plugin\skills\SKILL.md
+  - **Local copy**: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\teamwork_preview_auditor\android-cli_SKILL.md
+  - **Core methodology**: Android CLI task orchestration
+- **Source**: C:\Users\naina\.gemini\config\skills\github\SKILL.md
+  - **Local copy**: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\teamwork_preview_auditor\github_SKILL.md
+  - **Core methodology**: GitHub interaction and Git operations

package/.agents/teamwork_preview_auditor/ORIGINAL_REQUEST.md

@@ -0,0 +1,22 @@
+## 2026-06-25T13:49:23Z
+You are the Forensic Auditor for the Aether AI CLI project.
+Your working directory is C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\teamwork_preview_auditor
+Your identity is: teamwork_preview_auditor
+
+Your task is to audit the entire project codebase (`C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli`) for code integrity:
+1. Verify that no test results or expected output strings are hardcoded in the codebase to make tests pass.
+2. Verify that there are no dummy or facade implementations (e.g. mock endpoints or dummy CLI logic) that pretend to implement functionality without actual logic.
+3. Verify that the file context parsing, fallback calculations, API mesh routing, and CLI UX updates are implemented cleanly and authentically.
+4. Report your final audit verdict: CLEAN or INTEGRITY VIOLATION.
+5. List the specific checks performed and evidence gathered.
+
+Write a handoff report in your directory (handoff.md) and send a message back to the orchestrator (conversation ID: 94112169-cc09-4e27-b4f1-54773d8a3027) with your verdict and findings.
+
+## 2026-06-25T13:51:45Z
+You are the Victory Auditor for the Aether AI CLI project. The project orchestrator has claimed completion.
+Your task is to verify this claim independently.
+Workspace root: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli
+Orchestrator's Handoff: C:\Users\naina\.gemini\antigravity\scratch\aether-ai-cli\.agents\orchestrator\handoff.md
+Please conduct a rigorous 3-phase victory audit (timeline verification, cheating/facade/mock check, independent test suite execution) to determine if all requirements and acceptance criteria in ORIGINAL_REQUEST.md have been met.
+Deliver a structured verdict of either VICTORY CONFIRMED or VICTORY REJECTED, along with your audit findings report.
+

package/.agents/teamwork_preview_auditor/android-cli_SKILL.md

@@ -0,0 +1,203 @@
+---
+name: android-cli
+description: Orchestrates Android development tasks including project creation, deployment, SDK management, and environment diagnostics using the `android` command-line tool.
+---
+# Android CLI Specialist
+
+This skill provides instructions for using the `android` CLI tool. The tool includes various commands for creating projects, running applications, interacting with devices, and managing the CLI environment.
+
+## Installation
+
+If the android tool is not in the path install it. To install run the following command:
+Linux: curl -fsSL https://dl.google.com/android/cli/latest/linux_x86_64/install.sh | bash
+Mac: curl -fsSL https://dl.google.com/android/cli/latest/darwin_arm64/install.sh | bash
+Windows: curl.exe -fsSL https://dl.google.com/android/cli/latest/windows_x86_64/install.cmd -o "%TEMP%\i.cmd" && "%TEMP%\i.cmd"
+
+## SDK management
+To manage the installation of Android SDKs and tools, use the `sdk` command. For example:
+
+- `android sdk install <package>[@<version>]...`: Install specific packages. Multiple packages can be specified, separated by spaces. `<version>` defaults to latest. For example: `android sdk install platforms/android-30@2 platforms/android-34`
+- `android sdk update [<pkg-name>]`: Update a specific package or all packages to the latest version.
+- `android sdk remove <pkg-name>`: Remove a package from the local SDK.
+- `android sdk list --all`: List installed and available SDK packages.
+
+## Project creation
+Create projects from templates using the `create` command.
+
+For example: `android create empty-activity --name="My App" --output=./my-app`
+
+## Interacting with devices
+For more information on interacting with running devices, see [here](references/interact.md)
+
+## Running journey tests
+For more information on running journeys, see [here](references/journeys.md)
+
+## Doc searching
+The `docs` command searches authoritative, high-quality Android developer documentation in the Android Knowledge Base.
+By providing a few keywords, this tool will return high quality articles that contain examples or guidance on how to use Android APIs or libraries.
+Use this tool to obtain additional information on how to achieve Android-specific tasks or to know more about Android APIs, surfaces, libraries, or devices.
+
+Always use this tool to get the most up-to-date information about Android concepts. Typical good use cases are:
+  - Finding migration guides for APIs.
+  - Finding examples for APIs.
+  - Finding up-to-date information about Android APIs.
+  - Finding best practices for Android concepts.
+
+## Running APKs
+Use the `run` command to run Android apps.
+
+## Managing emulators
+
+Manage Android Virtual Devices (AVDs) using the `android emulator` command
+
+## Capturing screenshots
+
+Capture an image of the current screen of a connected Android device and output it to a file using the `android screenshot` command.
+
+## Managing skills
+
+Manage antigravity agent skills for Android using the `android skills` command.
+
+## Inspecting UI Layouts
+
+Use the `android layout` command to inspect the UI layout of an Android application. It returns the layout tree of an Android application in JSON format. When debugging UI errors, this is often a much faster approach than taking a screenshot.
+
+## Updating the CLI
+
+Update the Android CLI using the `android update` command.
+
+# `android help` output
+
+Usage: android [-hV] [--sdk=PARAM] [COMMAND]
+  -h, --help        Show this help message and exit.
+      --sdk=PARAM   Path to the Android SDK
+  -V, --version     Print version information and exit.
+Commands:
+  create    Create a new Android project
+  describe  Analyzes an Android project to generate descriptive metadata.
+  docs      Android documentation commands
+  emulator  Emulator commands
+  help      Shows the help of all commands
+  info      Print environment information (SDK Location, etc.)
+  init      Initializes the environment (eg. skills) for Android CLI.
+  layout    Returns the layout tree of an application
+  run       Deploy an Android Application
+  screen    Commands to view the device
+  sdk       Download and list SDK packages
+  skills    Manage skills
+  update    Update the Android CLI
+
+create
+          Usage: android create [-h] [--verbose] [--list] [--minSdk=api]
+                                --name=applicationName [-o=dest-path] [template-name]
+          Create a new Android project
+                [template-name]      The template name
+            -h, --help               Show this help message and exit.
+                --minSdk=api         The 'minSdk' supported by the application (default
+                                       is defined in the template)
+                --name=applicationName
+                                     The name of the application (e.g. 'My Application')
+            -o, --output=dest-path   The destination project directory path (default is
+                                       '.')
+                --verbose            Enables verbose output
+                --list               List all available templates
+
+describe
+          Usage: android describe [-hV] [--project_dir=PARAM]
+          Analyzes an Android project to generate descriptive metadata.
+          This command identifies and outputs the paths to JSON files that detail the
+          project's structure, including build targets and their corresponding output
+          artifact locations (e.g., APKs). This information enables other tools and
+          commands to locate build artifacts efficiently.
+            -h, --help                Show this help message and exit.
+                --project_dir=PARAM   The project directory to describe
+            -V, --version             Print version information and exit.
+
+docs
+          Usage: android docs [-h] [COMMAND]
+          Android documentation commands
+            -h, --help   Show this help message and exit.
+          Commands:
+            search  Search Android documentation
+            fetch   Fetch Android documentation
+
+emulator
+          Usage: android emulator [-h] [COMMAND]
+          Emulator commands
+            -h, --help   Show this help message and exit.
+          Commands:
+            create  Creates a virtual device
+            start   Launches the specified virtual device. This command will return when
+                      the emulator is fully started and ready to use.
+            stop    Stops the specified virtual device
+            list    Lists available virtual devices
+            remove  Delete a virtual device
+
+help
+          Usage: android help [COMMAND]
+          Shows the help of all commands
+                [COMMAND]   The command to show help for
+
+info
+          Usage: android info <field>
+          Print environment information (SDK Location, etc.)
+                <field>   The specific field to print the value of. If omitted print all.
+
+init
+          Usage: android init
+          Initializes the environment (eg. skills) for Android CLI.
+
+layout
+          Usage: android layout [-dhp] [--device=PARAM] [-o=PARAM]
+          Returns the layout tree of an application
+            -d, --diff           Returns a flat list of the layout elements that have
+                                   changed since the last invocation of ui-dump
+                --device=PARAM   The device serial number
+            -h, --help           Show this help message and exit.
+            -o, --output=PARAM   Writes the layout tree to the specified file or
+                                   directory. If omitted, prints the tree to standard
+                                   output
+            -p, --pretty         Pretty-prints the returned JSON
+
+run
+          Usage: android run [-h] [--debug] [--activity=PARAM] [--device=PARAM]
+                             [--type=PARAM] [--apks=PARAM[,PARAM...]]...
+          Deploy an Android Application
+                --activity=PARAM   The activity name
+                --apks=PARAM[,PARAM...]
+                                   The paths to the APKs
+                --debug            Run in debug mode
+                --device=PARAM     The device serial number
+            -h, --help             Show this help message and exit.
+                --type=PARAM       The component type (ACTIVITY, SERVICE, etc.)
+
+screen
+          Usage: android screen [-h] [COMMAND]
+          Commands to view the device
+            -h, --help   Show this help message and exit.
+          Commands:
+            capture  Outputs the device screen to a PNG
+            resolve  Target UI elements visually
+
+sdk
+          Usage: android sdk [COMMAND]
+          Download and list SDK packages
+          Commands:
+            install  Install SDK packages
+            update   Update one or all packages to the latest version
+            remove   Remove a package from the SDK
+            list     List installed and available SDK packages
+
+skills
+          Usage: android skills [COMMAND]
+          Manage skills
+          Commands:
+            add     Install a skill
+            remove  Remove a skill
+            list    List available skills
+            find    Find skills by keyword
+
+update
+          Usage: android update [--url=PARAM]
+          Update the Android CLI
+                --url=PARAM   The URL to download the update from

package/.agents/teamwork_preview_auditor/github_SKILL.md

@@ -0,0 +1,58 @@
+---
+name: github
+description: Interact with GitHub repositories, issues, pull requests, branches, commits, and workflows using the GitHub MCP server and Git CLI.
+---
+# GitHub Specialist Skill
+
+This skill guides the agent in using the GitHub MCP server and local Git CLI to interact with remote repositories on GitHub.
+
+## When to Use
+Use this skill when the user asks to:
+- Retrieve repository information, issues, or pull requests from GitHub.
+- Search code or repositories on GitHub.
+- Create, comment on, or update issues/PRs.
+- Inspect branch information, commits, or run workflows.
+
+## Authentication
+Ensure the GitHub MCP server is configured with a valid Personal Access Token (PAT) under the `GITHUB_PERSONAL_ACCESS_TOKEN` environment variable in `mcp_config.json`.
+
+## Standard Tools & Operations
+The GitHub MCP server provides the following categories of operations:
+
+### 1. Repositories
+- `get_repo(owner, repo)`: Fetch repository details.
+- `search_repos(query)`: Find repositories by query.
+
+### 2. Issues & Comments
+- `list_issues(owner, repo, state, labels)`: List issues.
+- `get_issue(owner, repo, number)`: Retrieve details of a specific issue.
+- `create_issue(owner, repo, title, body, assignees, labels)`: Create a new issue.
+- `update_issue(owner, repo, number, state, title, body, assignees, labels)`: Update issue properties.
+- `add_issue_comment(owner, repo, number, body)`: Add a comment to an issue.
+
+### 3. Pull Requests
+- `list_pull_requests(owner, repo, state)`: List pull requests.
+- `get_pull_request(owner, repo, number)`: Retrieve PR details.
+- `create_pull_request(owner, repo, title, head, base, body)`: Create a new pull request.
+- `merge_pull_request(owner, repo, number, commit_title, commit_message, merge_method)`: Merge a pull request.
+
+### 4. Code & Files
+- `get_file_contents(owner, repo, path, ref)`: Fetch the content of a file in the repo.
+- `create_or_update_file(owner, repo, path, message, content, branch, sha)`: Create or update a file.
+- `search_code(q)`: Search code across the repository.
+
+### 5. Branches & Commits
+- `list_branches(owner, repo)`: List branches.
+- `get_branch(owner, repo, branch)`: Get details of a branch.
+- `create_branch(owner, repo, branch, ref)`: Create a new branch.
+- `list_commits(owner, repo, sha, path, author, since, until)`: List commits.
+- `get_commit(owner, repo, sha)`: Get details of a specific commit.
+
+### 6. Workflows & Actions
+- `list_workflows(owner, repo)`: List workflows.
+- `list_workflow_runs(owner, repo, workflow_id)`: List runs of a workflow.
+- `get_workflow_run(owner, repo, run_id)`: Retrieve detailed run status.
+
+## Best Practices
+- **Read-Only Safeties**: When executing destructive actions or merging PRs, verify code changes first and ensure you have user consent.
+- **Search Optimization**: Use specific labels and states when querying issues to avoid fetching too many results and cluttering the context window.