Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Nagarjuna2997/ios-agent-skillWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/nagarjuna2997/ios-agent-skill/ios-build)<a href="https://agentmods.dev/commands/nagarjuna2997/ios-agent-skill/ios-build"><img src="https://agentmods.dev/badge/commands/nagarjuna2997/ios-agent-skill/ios-build/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/nagarjuna2997/ios-agent-skill/ios-build"><img src="https://agentmods.dev/badge/commands/nagarjuna2997/ios-agent-skill/ios-build.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00019 | $0.01744 |
| Opus 5.5 | $0.00008 | $0.00698 |
| Sonnet 5.5 | $0.00004 | $0.00349 |
| Haiku 4.5 | $0.00002 | $0.00174 |
Grade A, and why
ios-build scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 44 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/ios-build
Arguments: $ARGUMENTS
You are the iOS build agent. Turn the request into a working native SwiftUI app using the ios-agent MCP server tools (ios_preflight, ios_capabilities, ios_plan, ios_create_project, ios_add_capabilities, ios_write_files, ios_add_package, ios_build, ios_run, ios_screenshot, ios_design_evidence, ios_logs, ios_progress, ios_report). If those tools are not available, stop and tell the user to connect the ios-agent MCP server, from a build that exposes ios_design_evidence, and restart the session (this visual loop is not yet published to npm). Setup steps: https://github.com/Nagarjuna2997/ios-agent-skill/blob/main/docs/tooling/ios-build-agent.md#install
The user may not be a developer. Ask the user to choose a design direction before writing app code; also ask if a step needs credentials or money; make reasonable choices and state them as plan assumptions.
Status lines
Before each stage, print one short line (for example Planning screens and capabilities, Applying capability Sign in with Apple, Build attempt 2 of 8: fixing 3 errors, Launching in iPhone 17 Pro, Screenshot 2 of 3: Settings) and record the same line with ios_progress so it appears in RUN_REPORT.md.
Arguments
--out DIR: project folder (absolute, or relative to the current directory; make it absolute). Default:./ios-agent-apps/<kebab-case app name>.--resume: readDIR/.ios-agent/state.jsonandDIR/.ios-agent/plan.json, report where the run stopped, and continue from that stage. Do not redo finished stages.--refine "<change>": apply the change to the app inDIR(callios_progresswith stagegeneratingand the change), add any needed capabilities withios_add_capabilities, edit files withios_write_files, then rebuild, relaunch, re-screenshot and rewrite the report. On the refinement's firstios_build, passnewCycle: trueandchangeset to the change: the refinement gets its own attempt and time budget, and the change is listed in the report.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 6d ago Changed · +1 lines 7f9e989eb7ce
- 13d ago First seen · 43 lines · 19 tokens per session scan A d50ef5bf9dba
ios-build is a command published in the GitHub repository Nagarjuna2997/ios-agent-skill (38 stars, last pushed today), licensed MIT. It adds 19 tokens to every session and 1,744 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-03.
Other commands, from other repositories
feature-learn
View and manage patterns learned from feature builds. Shows extracted instincts, confidence scores, and feature completeness history.
feature-build
Build a complete mobile feature from description to running code with E2E tests. Orchestrates 6 phases - planning, implementation, testing, build-fix, quality gate, and verification. Auto-detects platform.
feature-plan
Plan a mobile feature with architecture decisions, file breakdown, dependency analysis, and test strategy. Produces a structured plan document reviewed by an architecture agent.
feature-implement
Implement a planned mobile feature using parallel specialized agents for network, data, architecture, UI, and wiring layers. Requires an approved feature plan.
feature-status
Show the current status of a feature build - which phase is active, what has completed, what remains, and any blockers.
feature-test
Create all tests for an implemented feature - unit tests (ViewModel, UseCase, Repository), Compose UI tests, and E2E user flow tests.