iCoderAgent: AI Code Assistant
Not in any tracked top chart right now.
Updated 19 Oct 2026 · rankings refresh through the day
DeveloperMahesh Shinde
CategoryDeveloper Tools
PriceFree
Released
Rating★ 0.0 (0)
Version1.0.11
Age rating4+
First seen by us
5 Oct 2026
5 Oct 2026Rating
0.0 ★
0 ratings
Measured
iCoderAgent: AI Code Assistant on the App Store
App details
App ID
6779451649
Publisher
Mahesh Shinde
On the store since
Jul 2026
Under a year
Minimum OS
iOS 16.4+
iPhone 8 and later
Devices
iPhone · iPad
Universal purchase
Content rating
4+
Languages
1
EN
Size
219 MB
Also listed in
Productivity
Where it ranks now
Every top chart it appears in across 30 countries.
| Country | Charts | Rank |
|---|---|---|
| Not charting right now. | ||
Ratings
App Store, worldwide.
Rating
0.0 ★
0 ratings
Measured
Ratings
0
Measured
About iCoderAgent: AI Code Assistant
iCoderAgent is a mobile AI coding assistant for iPhone and iPad.It Build software on move with iCoderAgent—your personal AI software engineer. Generate code, debug errors, explain complex logic, refactor projects, create APIs, write SQL, and develop apps directly from your iPhone or iPad using your favourite AI model.
Bring Your Own AI : Choose the model that works best for you. Your API keys remain under your control. They are securely stored in the iOS Keychain and sent directly to the AI provider you choose.
Compatible with
1. Anthropic Claude
2. OpenAI
3. Google Gemini
4. xAI Grok
5. DeepSeek
6. Kimi
7. Hugging Face
8. OpenRouter
9. Ollama
10. Apple MLX-LM
Benefits of iCoderAgent:
- Generate production-ready code
- Debug and fix bugs
- Explain any codebase
- Build REST APIs
- Write SQL
- Create documentation
- Build websites
- Support for Swift, Python, JavaScript, TypeScript, SQL, Java, C++, HTML, CSS and more
iCoderAgent runs an embedded Python interpreter on-device with a sandboxed tool system, so the agent can read and write project files, run database queries, search code, and use version control within an isolated workspace. An optional autonomous build loop can plan, execute, and verify a task end to end. A companion macOS application can optionally run builds and tests on the user's own computer.
iCoderAgent Unique Values :
- Runs natively on iPhone and iPad
- Bring Your Own AI provider Key
- Embedded Python execution
- Autonomous coding workflows
- Local project workspace with folder access
- Optional macOS, windows, linux companion for more complex projects
- Multiple LLM AI providers support
- Privacy-first design
Subscriptions provide a monthly quota of "conversations" (one query plus one response) and chat-history retention; users still pay their AI provider directly for model usage. The free plan now covers one single-page web project; subscriptions unlock the other project types, folders and connecting a laptop or cloud host. This apps is offering following auto-renewable subscriptions.
1. Scale - 5000 conversations/month,1000 chat history, BYOK AI Model
- Title of auto-renewing subscription : Scale
- Length of subscription - Monthly auto renewal subscription
- App privacy policy - https://aidigiapps.com/icoderagent/privacy.html and Terms of Use (EULA)
- https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
2. Growth - 1550 conversations/month,300 chat history, BYOK AI Model
- Title of auto-renewing subscription : Growth
- Length of subscription - Monthly auto renewal subscription
- App privacy policy - https://aidigiapps.com/icoderagent/privacy.html and Terms of Use (EULA)
- https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
3. Starter - 500 conversations/month,100 chat history, BYOK AI Model
- Title of auto-renewing subscription : Starter
- Length of subscription - Monthly auto renewal subscription
- App privacy policy - https://aidigiapps.com/icoderagent/privacy.html and
Bring Your Own AI : Choose the model that works best for you. Your API keys remain under your control. They are securely stored in the iOS Keychain and sent directly to the AI provider you choose.
Compatible with
1. Anthropic Claude
2. OpenAI
3. Google Gemini
4. xAI Grok
5. DeepSeek
6. Kimi
7. Hugging Face
8. OpenRouter
9. Ollama
10. Apple MLX-LM
Benefits of iCoderAgent:
- Generate production-ready code
- Debug and fix bugs
- Explain any codebase
- Build REST APIs
- Write SQL
- Create documentation
- Build websites
- Support for Swift, Python, JavaScript, TypeScript, SQL, Java, C++, HTML, CSS and more
iCoderAgent runs an embedded Python interpreter on-device with a sandboxed tool system, so the agent can read and write project files, run database queries, search code, and use version control within an isolated workspace. An optional autonomous build loop can plan, execute, and verify a task end to end. A companion macOS application can optionally run builds and tests on the user's own computer.
iCoderAgent Unique Values :
- Runs natively on iPhone and iPad
- Bring Your Own AI provider Key
- Embedded Python execution
- Autonomous coding workflows
- Local project workspace with folder access
- Optional macOS, windows, linux companion for more complex projects
- Multiple LLM AI providers support
- Privacy-first design
Subscriptions provide a monthly quota of "conversations" (one query plus one response) and chat-history retention; users still pay their AI provider directly for model usage. The free plan now covers one single-page web project; subscriptions unlock the other project types, folders and connecting a laptop or cloud host. This apps is offering following auto-renewable subscriptions.
1. Scale - 5000 conversations/month,1000 chat history, BYOK AI Model
- Title of auto-renewing subscription : Scale
- Length of subscription - Monthly auto renewal subscription
- App privacy policy - https://aidigiapps.com/icoderagent/privacy.html and Terms of Use (EULA)
- https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
2. Growth - 1550 conversations/month,300 chat history, BYOK AI Model
- Title of auto-renewing subscription : Growth
- Length of subscription - Monthly auto renewal subscription
- App privacy policy - https://aidigiapps.com/icoderagent/privacy.html and Terms of Use (EULA)
- https://www.apple.com/legal/internet-services/itunes/dev/stdeula/
3. Starter - 500 conversations/month,100 chat history, BYOK AI Model
- Title of auto-renewing subscription : Starter
- Length of subscription - Monthly auto renewal subscription
- App privacy policy - https://aidigiapps.com/icoderagent/privacy.html and
Latest updates
1.0.11
1.0.11 (31) — what users get over build 30
1. iCoderAgent is supporting iPhone Duo.
2. Spec interview, fixed properly. It was a sheet squeezed by everything else on screen, so the answers weren’t visible and the only thing you could do was tap Next — interviews came back with every question unanswered. It’s now full screen, and each question shows its suggested options plus an empty field so you can answer in your own words instead of only what the model guessed.
3. Task list is readable. A spec run makes 30+ items; they all rendered at once. Now 8 per page with page numbers, auto-turning to whatever the agent is working on. And the agent reconciles the list when a turn ends rather than leaving it part-ticked with no way to tell finished from abandoned.
4. Profile photos. Pick one in Profile; it shows there, on the Settings row, and on your own chat bubble with a speech tail. Stored per-account on the device, never uploaded.
5. Bigger agent image on big screens. The face and layout are 50% larger on iPad and iPhone Duo (open and closed), and the Duo header sits beside the side rail instead of fighting it.
6. Git actually explains itself. --force and -u were being read as remote names, and a rejected push printed two raw hashes. Flags work; failures say what’s wrong and what to do.
7. Audio. Ducking moved off the main thread — no more UI freeze warning, and the orange !pri error you saw twice is handled.
8. Not user-visible: Xcode 26.3 build fixes, the DEBUG “Unlock everything” toggle (compiled out of Release), and the macOS release-script fixes.
1.0.10
iCoderAgent 1.0.10 — What’s New
YOUR AGENT HAS A FACE AND A VOICE
Meet six characters: Robot, Owl, Penguin, Fox, Dragon and the original Droid. The five new mascots have full bodies, hands, feet and accessories, with blinking, head tilts, talking, waving and walking animations. Their speaking animations follow your voice.
Choose a character and your agent takes its name. Pick Fox, and your agent is Fox. This replaces the free-text name field.
Double-tap a character to cycle through its animations. The original Droid cycles through its twelve expressions. Explore every animation side by side in the new Settings gallery.
Enjoy spoken replies, voice dictation and a voice picker. Optional idle check-ins have a daily cap. Always-listening mode is available and remains off by default.
Choose how your agent replies: Plain, Detailed or Technical.
CHANGE IMPACT ANALYSIS
Before editing existing code, the agent checks four areas: scope, dependencies, risk and tests.
The new change_impact tool identifies callers left behind after a rename or deletion, including references to functions, classes, SQL tables and columns. It catches cases that find_references cannot detect once the old name has disappeared from the code tree.
Signature changes are highlighted too, helping identify callers affected even when the name stays the same.
Changes are classified as dependency, schema, contract, configuration, infrastructure, test-only or documentation-only. Each classification states the validation it requires.
Two safeguards are enforced:
• Destructive migrations are refused without a rollback.
• A task cannot be marked Done while tests are failing for that exact version of the code.
Reports are saved in .icoder/cia/reports/ and record what the analysis could not inspect. Rules in .icoder/cia/policy.json can make these safeguards stricter, but cannot weaken them.
SMARTER TESTING
Run tests related to your changes using run_tests related_to=. If no tests reference the changed code, the result explicitly says NOT COVERED.
JavaScript tests now run directly on your phone using the Node runtime inside the app, without a connected computer.
The agent now writes tests for logic it creates.
Fixed an issue where a Python project containing package.json could incorrectly be sent to the Node test runner.
GIT IMPROVEMENTS
Support for the full git remote subcommand set, plus branch renaming and deletion.
New repositories created with git init start on main.
Authentication tokens are kept out of .git/config, and credentials are scrubbed from all Git output.
SUBSCRIPTION FEATURES
Character faces, voice, microphone features and idle chat require a subscription. Locked controls explain why the feature is unavailable and take you directly to the subscription plans.
1.0.9
Talk to the agent without typing "agent" first — just say what you want to build. The agent now has an animated face that shows what it's doing, and you can give it a name in Settings. Every provider's API key field links straight to where you get the key. Added AirLLM so you can use a large model running on your own GPU over Wi-Fi, and fixed local models being cut off mid-reply. Markdown formatting in notebooks now applies properly.
What iOS users will notice
1. Just say what you want. No more typing agent first — a sentence goes to the agent, ls and git status still run as commands. agent … still works.
2. The agent has a face and a name. An animated avatar shows whether it's thinking, working or waiting on you. Name it in Settings and it answers to that name.
3. Get API key links. Every provider's key field now links straight to that provider's console, and Settings says plainly that you pay the provider directly.
4. AirLLM support. Point the app at a big model running on your own GPU machine over the LAN, alongside the existing Ollama and MLX-LM options.
5. Markdown toolbar works in notebooks. Bold, italic and headings applied and then vanished; they now stick.
6. Local models stop timing out. A model running on your own hardware is given time to answer instead of being cut off at two minutes mid-reply.