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/bentleypark/claude-code-mobile-spineWrote 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/bentleypark/claude-code-mobile-spine/feasibility)<a href="https://agentmods.dev/commands/bentleypark/claude-code-mobile-spine/feasibility"><img src="https://agentmods.dev/badge/commands/bentleypark/claude-code-mobile-spine/feasibility/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/bentleypark/claude-code-mobile-spine/feasibility"><img src="https://agentmods.dev/badge/commands/bentleypark/claude-code-mobile-spine/feasibility.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 | $0.00030 | $0.05574 |
| Opus 5.5 | $0.00012 | $0.02230 |
| Sonnet 5.5 | $0.00006 | $0.01115 |
| Haiku 4.5 | $0.00003 | $0.00557 |
Grade A, and why
feasibility 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 4d 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 — 218 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/mobile-spine:feasibility — Pre-design feasibility review
Every other spine path ends in a _tasks spec. This one stops before design is settled: it starts from a planning document alone — a prose spec, or an HTML prototype that stands in for one — and answers, before anyone builds a screen: does each platform allow what this document asks for, and what would it take in our repos? The output is one review document for planners and designers — _context/feasibility/{feature}.md.
The main session runs this command and does the extraction and the merge itself. Two kinds of judgment are delegated, in parallel:
| Layer | Who | Tools | Answers |
|---|---|---|---|
| OS / store policy | docs-agent |
WebFetch, WebSearch only | Does the OS allow it, from which version, under what policy — with fetched citations |
| Repo | android-agent / ios-agent §Feasibility check mode |
their usual tools, read-only mode | Version floor, what is declared, what already exists |
Neither layer sees the other's half: docs-agent cannot read the repos, and the platform agents have no WebFetch / WebSearch and do no web research in this mode. The verdict that reaches planners is the merge of the two.
User argument: $ARGUMENTS
Workspace prerequisite
Same checks as /mobile-spine:feat §Workspace prerequisite: .claude/mobile-spine.config.yaml must exist and pass the schema check (pm-agent.md Configuration → §Config schema versioning). Abort with the same messages if not, prefixed [/feasibility] instead of [/feat].
Configuration: read app from the config. In this file myapp stands for that value — ../myapp-android/ is ../{app}-android/, and so on.
Step 1 — Collect the inputs (ask only for what is missing)
Parse $ARGUMENTS first; ask only for what it does not supply, batched into one AskUserQuestion.
- Planning document — a path. Read it by type:
.md/.txt/.pdf→Readdirectly..docx→ convert to text:textutil -convert txt -stdout {path}(macOS), elsepandoc -t plain {path}. If neither is available, ask the user for a.pdfor.mdexport. Never guess at a document you could not read.- HTML prototype — a
.htmlfile, or a folder holding one (see §HTML prototype input below). - Check the text you read, whatever the format. A converter can succeed and still return nothing useful — a renamed
.doc, a protected file, requirements held in text boxes or pasted as images — and a long PDF read in page ranges can stop short. If the text is empty, far shorter than the document should be, or missing sections its own table of contents lists, try the other converter (or read the remaining pages), then ask for a PDF export. For a prototype, the check is that every screen its navigation or router names was found. Treat an input that still reads this way as unreadable (§Abort conditions). - Must not live inside
../myapp-android/,../myapp-ios/, or../myapp-backend/— ask the user to copy it into the workspace instead (platform repos are their agents' territory).
- Feature key — kebab-case, used for the output filename. Derive it from the document title when obvious and say so; ask only when it is not. For a prototype under
_context/design/{feature}/, the folder name is the key (when it is kebab-case; otherwise derive one and say so).
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.
- 4d ago First seen · 218 lines · 30 tokens per session scan A ef50a20585c4
feasibility is a command published in the GitHub repository bentleypark/claude-code-mobile-spine (1 stars, last pushed 5d ago), licensed MIT. It adds 30 tokens to every session and 5,574 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 2026-10-16.
Other commands, from other repositories
audit
Scan a codebase for existing observability instrumentation and identify gaps.
instrument
Generate a comprehensive instrumentation plan for a mobile codebase.
xcbuild
Build an iOS app for the iOS Simulator using xcodebuild with automatic project and simulator detection.
whattotest
Generate TestFlight WhatToTest document from changelog.
app-store-audit
Run an enterprise pre submission compliance audit on an iOS or Android app against Apple App Store and Google Play rejection rules. Pass a project path or run from the project root.
feature-learn
View and manage patterns learned from feature builds. Shows extracted instincts, confidence scores, and feature completeness history.