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.
npx skills add decocms/studio --skill flutter-appgit clone --depth 1 https://github.com/decocms/studioWrote 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/skills/decocms/studio/flutter-app)<a href="https://agentmods.dev/skills/decocms/studio/flutter-app"><img src="https://agentmods.dev/badge/skills/decocms/studio/flutter-app/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/skills/decocms/studio/flutter-app"><img src="https://agentmods.dev/badge/skills/decocms/studio/flutter-app.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 | $0.00078 | $0.03426 |
| Opus 5.5 | $0.00031 | $0.01370 |
| Sonnet 5.5 | $0.00016 | $0.00685 |
| Haiku 4.5 | $0.00008 | $0.00343 |
Grade A, and why
flutter-app scanned grade A with 1 finding 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 13d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
until curl -sfo /dev/null http://localhost:8099; do sleep 0.3; done # it binds after you ask How it starts
The opening of the file, as written. The whole thing — 268 lines — stays where its author put it; the contents beside it link to each section on GitHub.
flutter-app — building and QA'ing an app that is not a website
A Flutter repo breaks an assumption the rest of the sandbox is built on: its
pull requests have no deploy previewUrl, so the usual "open the preview and
look at it" does not apply — and neither does giving up.
First: check which sandbox image you are on
The Flutter SDK ships only in the Android sandbox image. If qa-android
is on your PATH (equivalently, $STUDIO_SANDBOX_ANDROID_EMULATOR is set),
you are on it: the SDK is at /opt/flutter (already on PATH), and there is
an Android emulator to run the app for real — see Running the app on the
emulator.
Otherwise you are on the default image, which has no Flutter SDK: no
flutter analyze, no flutter test, no way to see the UI. Do not try to
install the SDK into the sandbox. Do what reading the code can do, then say
in your report that you could not build or run the app and that the repo
must be switched to the Android emulator image in Settings → Repositories.
That is the fix; do not spend the run working around it.
Check the version pin first
flutter analyze produces different findings on different releases. Compare
the image's flutter --version against whatever the repo's CI pins (commonly
FLUTTER_VERSION in .github/workflows/*.yml, or .fvmrc). If they differ,
say so in your report and treat analyzer findings that the repo's own gate
would not produce as suspect. Do not "fix" them silently.
Private dependencies
Most real apps pull packages from private git repos:
dependencies:
some_internal_lib:
git:
url: https://github.com/acme/internal-flutter-lib.git
ref: v1.2.3
flutter pub get fetches those over plain HTTPS, and the sandbox is already set
up for it: the daemon installs the run's credentials in the pod's git config at
boot, so flutter pub get needs no preparation. Just run it.
Never run
gh auth setup-githere. The first thing it writes is an emptycredential.<url>.helper, which git reads as "discard every helper configured earlier" — and becauseGIT_CONFIG_GLOBALpoints at the daemon's config, that write lands in the same file and deletes the credentials the pod booted with. Apub getthat worked will start 404ing.
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.
- 13d ago Changed 2dec07ea3f66
- 22d ago Changed · +76 lines · +5 tokens per session 591ecfba18a6
- 23d ago Changed · +1 lines 32a41d2b4729
- 24d ago First seen · 191 lines · 73 tokens per session scan A b1f025fdc183
flutter-app is a skill published in the GitHub repository decocms/studio (409 stars, last pushed today), licensed MIT. It adds 78 tokens to every session and 3,426 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-22.
Other skills, from other repositories
swift-protocol-di-testing
Protocol-based dependency injection for testable Swift code — mock file system, network, and external APIs using focused protocols and Swift Testing.
credit-note-fixer
Fix the tiny credit-note formatting bug and rerun the exact targeted test command.
implement-universal
Harness-agnostic version of /implement. Drives a single workshop ticket through the SWE→Tester loop in ONE conversation, with the role prompts bundled as agents/software-engineer.md and agents/tester.md instead of being launched as subagents. Resolves the ticket from implementyourself/tasks/, creates an…
playwright-cli
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
verify-omnigent
Drive Omnigent the way a user does and prove a behavior with recorded evidence, using an isolated server, runner, and mock model plus a feature map of every user entry point. Load before reproducing a user-facing bug, before claiming a fix works, or when reviewing whether a change covered every surface (session vs.…
resolve-author-fix
Find the root cause, implement a focused fix, and prove behavior with targeted tests and recordings.