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 skills/fusebase-dev/fusebase-flow/client-vs-internalnpx skills add fusebase-dev/fusebase-flow --skill client-vs-internalgit clone --depth 1 https://github.com/fusebase-dev/fusebase-flowWhat 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 | $0.00104 | $0.01742 |
| Opus 5 | $0.00052 | $0.00871 |
| Sonnet 5 | $0.00021 | $0.00348 |
| Haiku 4.5 | $0.00010 | $0.00174 |
Grade A, and why
client-vs-internal 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 2d 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 — 124 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Client vs Internal
Style: Mode-B-lite. Artifact-gated by default — inert unless
docs/audience.mdexists or the operator explicitly requests the posture check.
Purpose
Client-facing teams build for two very different audiences. Clients need simplicity and trust (never Salesforce-complex); internal teams need robust controls and power-features. This skill applies the right posture to each surface, using the audiences defined at onboarding (or the audience the operator states when explicitly requesting the check).
When to invoke
docs/audience.mdexists AND a UI/app surface is being designed or built.- Operator says "is this for clients or internal", "simplify for the client", "this is an internal tool".
Do not invoke when
docs/audience.mdabsent and no explicit operator request → silent no-op; do not activate or create it.- Non-UI / backend-only work with no audience-facing surface.
Required inputs
| Input | Where it lives | If missing |
|---|---|---|
| Audience definitions | docs/audience.md or the audience stated in an explicit operator posture-check request |
STOP — no-op only when both are absent. Do not create the file; onboarding (/onboard) creates it. |
| Surface being built | current task | nothing to classify; exit |
Procedure
- Existence gate (FIRST STEP). No
docs/audience.mdand no explicit operator posture-check request → exit silently. If the operator explicitly requested the check and stated the audience, continue with that audience without creating the file. - Read
docs/audience.mdwhen present (which surfaces are client-facing vs internal; per-audience needs); otherwise use the audience stated in the explicit operator request. - Classify the surface in scope: client-facing, internal, or shared.
- Apply the matching posture checklist (§ Posture checklists below): client-facing, internal, or shared — walk every row for the surface in scope; each row is pass / fail / N-A, not vibes.
- Persist — the classification must not evaporate in chat. If a spec is in flight (or being drafted) for this surface, hand the result to
requirements-specification: the posture + each applicable checklist row becomes a numbered spec AC citing its ID (e.g., "AC7 — client-facing surface: destructive actions confirm with object + consequence before executing (client-vs-internal C2)"), alongside the QP-xx ACs fromapp-quality-patterns(C2/C4 overlap the QP delete-policy and empty/loading/error patterns — cite both IDs on one AC line, don't duplicate the AC). No spec in flight → record posture + failed rows in the design brief or change-note so Implement andcode-reviewinherit them. - Flag mismatches (e.g. exposing internal complexity to a client surface) and recommend the audience-appropriate alternative.
- Ambiguous audience → ask the operator (FR-19).
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.
- 2d ago First seen · 124 lines · 104 tokens per session scan A 614ac467d2a7
client-vs-internal is a skill published in the GitHub repository fusebase-dev/fusebase-flow (9 stars, last pushed 7d ago), licensed MIT. It adds 104 tokens to every session and 1,742 once invoked, about $0.0005 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 skills, from other repositories
lov-maintain-partners
Maintain the Skill Publisher website's partners section AND align partner logo rows on event posters / hero strips: reuse lov-find-logo for brand logo discovery, normalize collected logos to a 240px-tall content canvas (retina-ready), rasterize SVGs via rsvg-convert before normalizing (so SVG viewBox padding gets…
lov-bp-polish
Review and improve an existing BP outline, PPTX, PDF, or rendered slide set across investment logic, evidence, copy, charts, and visual quality. Produces a scored report, page-level revisions, and targeted regeneration instructions while keeping facts separate from assumptions. Trigger on "润色 BP", "审稿商业计划书", "PPT…
lov-business-card
Generate a professional, editorial-style business card (2:1) as a high-resolution PNG and a self-contained interactive HTML. Collects the user's identity (name, roles, tagline, interests, locations, optional avatar), fills a parameterized design template with a chosen theme, and renders it via headless Chrome. Trigger…
vibe-to-ui
Design systems, motion, mood boards, spatial layout, visual assets, Design Context profiles, and a global Inspiration Library from screenshots, website URLs, inspiration images, music, or fuzzy aesthetic intent. Classifies page archetype, explores 3 product-aware directions before locking tokens (unless exact…
design-improvement
WHAT - Browser-grounded iterative design improvement. Consumes design-assessment findings, defines direction, prioritizes safe vs ambiguous changes, implements within existing design system, runs app, captures rendered evidence via browser, reviews and iterates. Reuses evidence model — no new scoring framework.
design-assessment
WHAT - Evidence-based design-unit assessment orchestrated by project-assessment. Evaluates visual hierarchy, UX friction, interaction, a11y, responsiveness, design-system compliance, and distinctiveness with severity/confidence evidence citations. Reuses project-assessment-evidence semantics — no second framework.