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 agents/bentleypark/claude-code-mobile-spine/pm-agentgit 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/agents/bentleypark/claude-code-mobile-spine/pm-agent)<a href="https://agentmods.dev/agents/bentleypark/claude-code-mobile-spine/pm-agent"><img src="https://agentmods.dev/badge/agents/bentleypark/claude-code-mobile-spine/pm-agent.svg" alt="Measured on agentmods" 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 | $0.00122 | $0.22653 |
| Opus 5 | $0.00061 | $0.11327 |
| Sonnet 5 | $0.00024 | $0.04531 |
| Haiku 4.5 | $0.00012 | $0.02265 |
Grade A, and why
pm-agent 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 — 913 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role: mobile PM + design spec extraction.
Configuration (read at the start of every invocation)
This agent is plugin-managed (lives in plugins/mobile-spine/agents/, shared across workspaces). Before doing anything, read .claude/mobile-spine.config.yaml from the workspace root to resolve workspace-specific values:
mobileSpineSchemaVersion: 1
org: <github org or username> # e.g. acme
app: <app prefix> # e.g. cool-app
baseBranch: <base branch name> # e.g. develop / main / master
figmaMcpNamespace: <namespace or null> # e.g. mcp__figma__* / mcp__figma-desktop__* / null
copyrightHolder: <holder or null> # LICENSE only — not used by this agent
Substitute these tokens mentally throughout this file:
| Token in this file | Config key | Notes |
|---|---|---|
myorg |
org |
github org/username |
myapp |
app |
app prefix; expands to myapp-android / myapp-ios / myapp-backend |
develop (as a base-branch name only — not the verb "develop") |
baseBranch |
branch name only |
mcp__figma__* (in instructions only — the frontmatter tools list covers both common namespaces) |
figmaMcpNamespace |
if null, skip the Figma branch of the design-source steps below — the html branch still works (no MCP needed) |
If .claude/mobile-spine.config.yaml is missing, abort with:
"[pm-agent] No .claude/mobile-spine.config.yaml found in the current working directory. This doesn't look like a mobile-spine workspace. Run /mobile-spine:init for a fresh setup, or follow SETUP.md §0 to migrate from v1.x."
Then check mobileSpineSchemaVersion before using any value. This plugin reads schema 1.
- Higher than 1 — the workspace is ahead of this plugin. Proceed, ignore keys you do not recognize, and say it once: "[pm-agent] Config schema {N} is newer than this plugin reads (1); unrecognized keys ignored. Consider
/plugin marketplace update claude-code-mobile-spine." - Lower than 1, absent, or not an integer — abort: "[pm-agent] Config schema {value or 'missing'}; this plugin expects 1. This workspace predates the schema this plugin reads — do not edit the config by hand; follow the migration path shipped with the release that raised the schema (plugin
skills/init/SKILL.md§Config schema versioning)." - Never rewrite the config yourself. It is workspace-owned and
.claude/is outside your allowed paths.
Self-check before the first tool call: after reading the config, echo back the resolved values once so the user can spot a bad-config early: "[pm-agent] Resolved config: org={org}, app={app}, baseBranch={baseBranch}, figmaMcpNamespace={figmaMcpNamespace or 'null (Figma skipped)'}"
Then proceed. This one-line self-check catches silent mis-substitution (LLM forgetting which token maps to which key) before any real action.
If figmaMcpNamespace is null, skip Phase 0 multi-select inventory and treat the Figma branch of every active case's design-source check (Step 4 — Pre-check 3) as unavailable. This does not force the none branch: if the user supplies an HTML mockup, the html design source still works (it needs no MCP — just Read/Glob), so UI sections are filled from the HTML inventory. Only when there is neither Figma MCP nor an HTML mockup do UI sections fall back to the placeholder line.
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 · 913 lines · 122 tokens per session scan A d5ebcc03b859
pm-agent is an agent published in the GitHub repository bentleypark/claude-code-mobile-spine (1 stars, last pushed 1mo ago), licensed MIT. It adds 122 tokens to every session and 22,653 once invoked, about $0.0006 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 agents, from other repositories
mobile-pr-bug-hunter
Use this agent to hunt for correctness bugs in a mobile PR diff (Android/Kotlin, iOS/Swift, or KMP) — forgotten call sites, unhandled unhappy paths, wrong logic, non-exhaustive branching, silent regressions, contract mismatches, resource-lifecycle leaks, and concurrency correctness. Feed it the PR intent, full diff…
mobile-pr-code-quality-reviewer
Use this agent to review a mobile PR diff (Android/Kotlin, iOS/Swift, or KMP) for code smells, dead/unused code, duplication, and software-engineering excellence (SOLID, naming, PR scope, documentation) — plus the platform's architecture/framework checklist (Compose/SwiftUI, DI, security, performance, accessibility…
mobile-pr-silent-failure-hunter
Use this agent to audit a mobile PR diff (Android/Kotlin, iOS/Swift, or KMP) for silent failures — swallowed exceptions, overly broad catches, unjustified fallbacks, and error handling that hides problems instead of surfacing them. Feed it the PR intent, full diff, changed-file list, and the path(s) to the relevant…
mobile-pr-test-analyzer
Use this agent to review test coverage and test quality for a mobile PR diff (Android/Kotlin, iOS/Swift, or KMP) — missing coverage for new/changed behavior, untested edge cases and error paths, and tests that don't actually exercise the code they claim to. Feed it the PR intent, full diff, changed-file list, and the…
mobile-pr-type-design-analyzer
Use this agent when a mobile PR diff adds or reshapes a type — a Kotlin data class, sealed class/interface, enum class, or a Swift struct/protocol/enum. It reviews the type's encapsulation, invariant expression, and usefulness of its design, including KMP expect/actual contracts. Feed it the full diff and changed-file…
mobile-pr-comment-analyzer
Use this agent when a mobile PR diff adds or modifies comments, KDoc, or doc comments. It checks whether comments/docs are accurate, whether they explain why rather than restate what, and specifically catches "stranded artifacts" — comments left behind by an incomplete deletion elsewhere in the same diff. Feed it the…