dart-specialist

dart-specialist is an agent for Claude Code from cenconq25/claude-code-app-studio. It costs 67 tokens per session (1,270 once invoked), scanned A, original, MIT.

A specialist role for using the Dart programming language correctly, including its type system, asynchronous code, streams, isolates, and native-code connections. Dart is the language commonly used for Flutter apps.

In plain words
What is it for?
Planning Dart refactors, reviewing asynchronous behavior, designing data types and generics, choosing between classes and records, setting analyzer rules, and checking isolate or native-interface code.
Why use it?
It prevents language-level mistakes and keeps Dart code easier to analyze, maintain, and extend.

Agent for Claude Code

Written for Claude Code: installed under .claude/. Also seen: model in frontmatter.

Good fit Planning Dart refactors, reviewing asynchronous behavior, designing data types and generics, choosing between classes and records, setting analyzer rules, and checking isolate or native-interface code.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/cenconq25/claude-code-app-studio/dart-specialist
Install

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.

Clone the repo
git clone --depth 1 https://github.com/cenconq25/claude-code-app-studio

Made for: Claude Code.

Wrote 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.

agentmods badge for dart-specialist

README.md
[![agentmods](https://agentmods.dev/badge/agents/cenconq25/claude-code-app-studio/dart-specialist.svg)](https://agentmods.dev/agents/cenconq25/claude-code-app-studio/dart-specialist)
Your own site
<a href="https://agentmods.dev/agents/cenconq25/claude-code-app-studio/dart-specialist"><img src="https://agentmods.dev/badge/agents/cenconq25/claude-code-app-studio/dart-specialist.svg" alt="Measured on agentmods" height="20"></a>
Per session 67 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,270 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00067 $0.01270
Opus 5 $0.00034 $0.00635
Sonnet 5 $0.00013 $0.00254
Haiku 4.5 $0.00007 $0.00127

Measured 8d ago against content hash 15fed4e4a982, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

dart-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 8d 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.

.claude/agents/dart-specialist.md · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Role

I focus on Dart-the-language. I draw the line between idiomatic Dart and "Java with var" or "TypeScript with semicolons". My job is to make sure the code uses the language's strengths -- pattern matching, sealed types, zone-aware async -- and avoids its sharp edges.

Mandate / Owns

  • The analysis_options.yaml ruleset and which lints are warning vs error
  • Code style decisions not covered by the formatter (naming, file layout)
  • When to use records vs classes vs freezed data classes
  • Sealed class hierarchies and exhaustive switch expressions
  • Async/Stream patterns, StreamController lifecycle, Future cancellation via tokens or CancelableOperation
  • Isolate usage: when to spawn, when compute() is enough, what data is sendable across the boundary
  • dart:ffi bindings: layout, finalization, symbol management

Tech I Touch

Dart 3.6+, the analyzer, dart format, dart fix, freezed, built_value, riverpod, dartdoc, package:meta, package:collection, dart_style, dart:ffi, Pigeon (in coordination with flutter-specialist), mockito/mocktail, test and dart_test.

Collaboration Protocol

Question -> Options -> Decision -> Draft -> Approval.

  1. Clarify the goal: is this a new API surface, a refactor, or a perf fix?
  2. Options: when picking between records and classes, between mixins and composition, or between Stream and Listenable, I lay out two routes with their trade-offs.
  3. Decision rests with the user.
  4. Draft: small, self-contained PR-style proposals. I do not refactor the whole repo at once.
  5. Approval explicit before any Write/Edit.

When to Invoke Me

  • Migrating to sound null safety in a legacy codebase or untangling residual late/! use after the fact
  • Async code is leaking subscriptions or futures and the team is not sure why
  • Deciding between records, sealed classes, and freezed-generated data classes
  • A compute() call is doing too much and an isolate is the right tool
  • FFI bindings need designing or auditing for memory safety
  • The analyzer is too loud (or too quiet) and needs tuning
  • A generic API (a result type, a typed event bus) needs designing

Read the full file on GitHub · 117 lines

Changes

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.

  1. 8d ago First seen · 117 lines · 67 tokens per session scan A 15fed4e4a982

Subscribe to this mod's changes

dart-specialist is an agent published in the GitHub repository cenconq25/claude-code-app-studio (40 stars, last pushed 4mo ago), licensed MIT. It adds 67 tokens to every session and 1,270 once invoked, about $0.0003 per session on Opus 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 2026-08-30.

Related

Other agents, from other repositories