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 agentmods add commands/vmobifystudio/app-dev-team/app-initgit clone --depth 1 https://github.com/vmobifystudio/app-dev-teamWhat 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 | $0.00022 | $0.02799 |
| Opus 5 | $0.00011 | $0.01399 |
| Sonnet 5 | $0.00004 | $0.00560 |
| Haiku 4.5 | $0.00002 | $0.00280 |
Grade A, and why
app-init 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 yesterday.
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 — 187 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/app-init — Kick off a new app project
You are starting a fresh app project. The user's one-liner (if any) is:
$ARGUMENTS
Steps
-
Record the founder's words before interpreting them. Create
docs/00-founder-intent/(copy${CLAUDE_PLUGIN_ROOT}/docs/00-founder-intent/README.mdinto it), write the user's brief and any material they supplied there verbatim and dated, then record it:node "${CLAUDE_PLUGIN_ROOT}/scripts/founder-intent.mjs" --project-root . --writeThis happens first because everything below is an interpretation of it, and an interpretation with nothing to be checked against is what makes this team's loop closed. The directory is append-only: never edit a recorded file, add a dated line to
decisions.mdinstead. -
Run the
requirements-intakeskill to get a cleandocs/01-intake.md. If the user's prompt is already detailed, ask only the questions still unanswered.
1a. Activate the roster. Invoke the role-activation skill: fix the tier (--utility /
--flagship, else derived from the intake) and the product type (from the intake's
## Product type answer), then write docs/02-team-roster.md — copy
${CLAUDE_PLUGIN_ROOT}/docs/02-team-roster.md and fill it in from the matrix; every role in the
activation matrix, each
active / conditional / off with its trigger or reason.
If the product type is unstaffed (cli), refuse and stop here — print the skill's
ACTIVATION REFUSED block naming the missing IC role. Write no roster and spawn nobody. A team
that cannot build the product is not a lesser start, it is a sprint of stranded tickets.
Write it before spawning anyone. Every fan-out below spawns only active roles, and this
file is what says so; deriving activation again per step is how two steps come to disagree.
Print the tier, the product type, and the off-list with reasons.
- Spawn the
ceoagent with the intake and the roster as input. CEO writesdocs/00-vision.md. Onutilitytier the CEO runs the founder pass — vision plus the PRD and backlogcpowould have written — andcpois not spawned at all (its roster row already says why).
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.
- yesterday First seen · 187 lines · 0 tokens per session scan A f0920379202c
app-init is a command published in the GitHub repository vmobifystudio/app-dev-team (4 stars, last pushed 22d ago), licensed MIT. It adds 22 tokens to every session and 2,799 once invoked, about $0.0001 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 2026-08-31.
Other commands, from other repositories
feature
Post a feature to the Features Board on production and write tweets.
commit-all
Group all changes semantically and commit each group separately.
test-feature
Test a React Native feature on the running simulator/emulator. Verifies UI, user flows, and internal state. Generates a persistent Maestro test file.
factory-next
Recommend and optionally run the next factory stage for this repo.
android-build
Build Android project with Gradle, fix errors, generate APK/AAB. Invokes android-build-resolver for issues.
compose-preview
Generate and verify Compose previews for UI components.