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 MichelKerkmeester/skilled-agent-harness_spec-driven-loops --skill sk-code-mobile-cligit clone --depth 1 https://github.com/MichelKerkmeester/skilled-agent-harness_spec-driven-loopsWrote 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/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-mobile-cli)<a href="https://agentmods.dev/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-mobile-cli"><img src="https://agentmods.dev/badge/skills/michelkerkmeester/skilled-agent-harness_spec-driven-loops/sk-code-mobile-cli.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 5.1 | $0.00025 | $0.06873 |
| Opus 5 | $0.00013 | $0.03436 |
| Sonnet 5 | $0.00005 | $0.01375 |
| Haiku 4.5 | $0.00003 | $0.00687 |
Grade C, and why
sk-code-mobile-cli scanned grade C 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 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- Keywords: pi-remote, mobile-cli, app-mobile, app-relay, packages/pi-rpc-protocol, svelte, sveltekit, runes, scoped-styles, app.css, design-system, token-library, component-tokens, retint-recipes, theme-remap, commen How it starts
The opening of the file, as written. The whole thing — 455 lines — stays where its author put it; the contents beside it link to each section on GitHub.
mobile-cli Surface — Pi Remote Design-System and Svelte Evidence
Read-only evidence for code work on the Pi Remote Mobile-CLI app (app-mobile/ and its relay
peer app-relay/, with the shared protocol in packages/pi-rpc-protocol/). The shipped web surface is
a SvelteKit app with Svelte runes and component-scoped styles. When the hub bundles this surface, a
code workflow gains the app's formalized design system, source-tree conventions, natural inline-comment
grammar, guardrails that keep designer edits out of logic and the security boundary, and the
browser-free verification gate this codebase actually uses.
1. WHEN THE HUB BUNDLES THIS
- The task's CWD or changed/target files sit under
app-mobile/,app-relay/, orpackages/pi-rpc-protocol/in the Mobile CLI app repository. The hub's surface detection resolves PI_REMOTE. - The active workflow phase needs the app's design-system evidence: the primitive → semantic →
component token model, the natural inline-comment editability convention, the
Do not edit — <why>guardrail notes, or the browser-free verification gate. - The active workflow needs the shipped Svelte source grammar: kebab-case names, SvelteKit route exemptions, runes lifecycle rules, shared-folder ownership, section banners, or paired source-folder documentation.
- This surface never owns edits, tests, or verification itself. The bundled workflow mode does the
work (for example,
sk-code-qualityorsk-code-review) and this packet supplies the evidence it must honor. A typical resolution is[sk-code-quality, sk-code-mobile-cli]or[sk-code-review, sk-code-mobile-cli].
What ships with it
60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
- assets/a11y-parity-checklist.md 3.6 KB
- assets/bem-rename-checklist.md 4.8 KB
- assets/ds-verification-checklist.md 6.0 KB
- assets/guardrail-audit-checklist.md 4.8 KB
- assets/runes-effect-audit-checklist.md 4.1 KB
- assets/story-coverage-checklist.md 3.2 KB
- assets/token-retint-checklist.md 5.0 KB
- changelog/v0.1.0.0.md 4.3 KB
- changelog/v0.1.1.0.md 3.0 KB
- changelog/v0.1.10.0.md 3.6 KB
- changelog/v0.1.11.0.md 3.1 KB
- changelog/v0.1.2.0.md 1.8 KB
- changelog/v0.1.3.0.md 991 B
- changelog/v0.1.4.0.md 790 B
- changelog/v0.1.5.0.md 1.4 KB
- changelog/v0.1.6.0.md 2.5 KB
- changelog/v0.1.7.0.md 1.2 KB
- changelog/v0.1.7.1.md 818 B
- changelog/v0.1.8.0.md 3.3 KB
- changelog/v0.1.9.0.md 2.7 KB
- manual-testing-playbook/cross-cli-dispatch/large-prompt-stress.md 8.1 KB
- manual-testing-playbook/cross-cli-dispatch/multi-step-dispatch.md 7.7 KB
- manual-testing-playbook/cross-cli-dispatch/short-prompt-baseline.md 5.7 KB
- manual-testing-playbook/holdout/accessibility-natural.md 6.2 KB
- manual-testing-playbook/holdout/debugging-natural.md 6.0 KB
- manual-testing-playbook/holdout/implementation-natural.md 6.4 KB
- manual-testing-playbook/holdout/ind-code-quality.md 5.7 KB
- manual-testing-playbook/holdout/ind-language-standards.md 5.8 KB
- manual-testing-playbook/holdout/ind-verification.md 5.2 KB
- manual-testing-playbook/intent-detection/accessibility-routing.md 5.4 KB
- manual-testing-playbook/intent-detection/comment-convention-routing.md 6.0 KB
- manual-testing-playbook/intent-detection/debugging-routing.md 5.1 KB
- manual-testing-playbook/intent-detection/guardrail-routing.md 5.2 KB
- manual-testing-playbook/intent-detection/language-standards-routing.md 5.5 KB
- manual-testing-playbook/intent-detection/token-edit-routing.md 6.1 KB
- manual-testing-playbook/intent-detection/verification-routing.md 5.3 KB
- manual-testing-playbook/manual-testing-playbook.md 15 KB
- manual-testing-playbook/resource-loading/mixed-load.md 6.3 KB
- manual-testing-playbook/resource-loading/references-only-load.md 7.0 KB
- manual-testing-playbook/surface-detection/negative-control-non-mobile-cli.md 6.3 KB
- manual-testing-playbook/surface-detection/pi-remote-positive-detection.md 5.7 KB
- manual-testing-playbook/token-cost-baseline/ceiling-load-all.md 7.9 KB
- manual-testing-playbook/token-cost-baseline/floor-single-resource.md 5.1 KB
- manual-testing-playbook/token-cost-baseline/median-load.md 5.3 KB
- manual-testing-playbook/unknown-fallback/ambiguous-multi-intent.md 6.9 KB
- manual-testing-playbook/unknown-fallback/disambiguation-required.md 7.1 KB
- manual-testing-playbook/unknown-fallback/zero-keyword-prompt.md 5.8 KB
- README.md 7.4 KB
- references/conventions/comment-grammar.md 8.0 KB
- references/conventions/conventions.md 3.9 KB
- references/conventions/editability-guardrails.md 5.1 KB
- references/conventions/folder-docs.md 4.4 KB
- references/design-system/component-tokens.md 6.7 KB
- references/design-system/css-class-naming-bem.md 5.3 KB
- references/design-system/design-system.md 4.8 KB
- references/design-system/retint-recipes.md 7.4 KB
- references/design-system/scoped-style-ownership.md 5.6 KB
- references/design-system/theme-remap.md 7.2 KB
- references/design-system/token-library.md 5.1 KB
- references/operations/incident-playbooks.md 5.7 KB
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 · 455 lines · 25 tokens per session scan C 41d92d76b55a
sk-code-mobile-cli is a skill published in the GitHub repository MichelKerkmeester/skilled-agent-harness_spec-driven-loops (34 stars, last pushed today), licensed MIT. It adds 25 tokens to every session and 6,873 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
uniapp-app-generate-skill
This skill should be used when the user wants to create a standardized uni-app project for WeChat mini-program, H5, and App from scratch. It guides through pre-development brainstorming, project initialization with uni-app + Vue3 + TypeScript + Pinia, dynamic theme system setup, static asset generation, layout design…
capacitor-vue
Guides the agent through Vue-specific patterns for Capacitor app development. Covers Vue 3 Composition API with Capacitor plugins, custom composables for native features, reactive plugin state, lifecycle hook patterns, Vue Router deep link integration, platform detection, PWA Elements setup, Quasar Framework…
ionic-vue
Guides the agent through Ionic Vue development patterns — project structure, Vue-specific Ionic components (IonPage, IonRouterOutlet, IonTabs), navigation with Vue Router and useIonRouter, Ionic lifecycle hooks (onIonViewWillEnter, onIonViewDidEnter, onIonViewWillLeave, onIonViewDidLeave), composable utilities…
uview-pro
A component guide for uView Pro, a UI library for uni-app applications. uni-app is a Vue-based framework for building apps for mobile, web, and several mini-program platforms.
mobile-responsive
A guide for making Vue 3 and Element Plus administration pages work well on phones as well as desktop screens. Vue 3 is a JavaScript interface framework, and Element Plus is its UI component library.
frontend-specialist
Skill "frontend-specialist" from Prorise-cool/Claude-Code-Multi-Agent, covering frontend specialist, when to use this skill, capabilities, android sdk and angular.