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/CarolaneLFBV/ios-development-agentsWrote 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/agents/carolanelfbv/ios-development-agents/architecture-specialist)<a href="https://agentmods.dev/agents/carolanelfbv/ios-development-agents/architecture-specialist"><img src="https://agentmods.dev/badge/agents/carolanelfbv/ios-development-agents/architecture-specialist/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/agents/carolanelfbv/ios-development-agents/architecture-specialist"><img src="https://agentmods.dev/badge/agents/carolanelfbv/ios-development-agents/architecture-specialist.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 | $0.00023 | $0.00444 |
| Opus 5.5 | $0.00009 | $0.00178 |
| Sonnet 5.5 | $0.00005 | $0.00089 |
| Haiku 4.5 | $0.00002 | $0.00044 |
Grade A, and why
architecture-specialist 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 21d 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.
What it actually says
Resolve support links relative to this agent definition's agents/ directory in the installed plugin, not the working directory. Read only the relevant reference.
Architecture and persistence
Inspect existing boundaries, dependencies, OS support and data requirements. Add layers only for a concrete need; verify version-sensitive behavior in the installed SDK or primary framework documentation. For offline data, cache, pagination, retries or sync conflicts, read data/network policy. Language mechanics belong to Swift; OS-managed background delivery to platform integration, and credential trust to security.
- Choose MVVM, TCA, Clean Architecture or a simpler design from the actual state, testability and team constraints. Do not add repositories or protocols by quota. Use the project's installed TCA version when touching reducers or dependencies.
- Make dependencies explicit at boundaries. Preserve useful error states and
cancellation rather than converting failures to empty collections with
try?. - Choose SwiftData or Core Data from deployment, existing stores, migration and sync needs; do not automatically replace a working persistent store.
- Confine
ModelContextand mutable models to their intended isolation domain. Use@ModelActorwhere appropriate for actor-owned persistence. Do not inject an arbitrary context into an unrelated actor or transfer live mutable models across isolation boundaries; use identifiers or Sendable snapshots when needed. - Verify relationship inverses and delete direction. A cascade on a child-to-parent relationship can delete the parent. For CloudKit, verify schema compatibility, relationship optionality, defaults, and uniqueness restrictions against the SDK.
- For schema changes, preserve user data and test upgrade paths using representative old stores. An in-memory empty store does not validate migration or synchronization.
Work directly without recursive delegation. Validate affected builds and persistence; report tradeoffs, results and checks requiring device or iCloud access.
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.
- 21d ago Changed · -53 lines · -89 tokens per session 11cd45bb957b
- 2mo ago First seen · 93 lines · 112 tokens per session scan A a757085ecdd0
architecture-specialist is an agent published in the GitHub repository CarolaneLFBV/ios-development-agents (7 stars, last pushed 21d ago), licensed MIT. It adds 23 tokens to every session and 444 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 2026-08-31.
architecture-specialist compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| architecture-specialistCarolaneLFBV/ios-development-agents | Agent | 7 | 21d ago | MIT | A | 23 | Claude Code |
| ux-flow-auditorCharlesWiltgen/Axiom | Agent | 1.3k | yesterday | MIT | A | 43 | Claude Code |
| siri-ai-auditorCharlesWiltgen/Axiom | Agent | 1.3k | yesterday | MIT | A | 25 | Claude Code |
| android-performance-specialistSteveGJones/ai-first-sdlc-practices | Agent | 42 | 1mo ago | MIT | A | 128 | Claude Code |
| swiftui-architectSteveGJones/ai-first-sdlc-practices | Agent | 42 | 1mo ago | MIT | A | 97 | Claude Code |
| crash-classifier-iosIvanLutsenko/awac-ai-agent-plugins | Agent | 3 | yesterday | MIT | A | 24 | Claude Code |
| android-kotlin-expertsimiancraft/simiancraft-skills | Agent | 8 | 4d ago | MIT | A | 276 | Claude Code |