Gamma Cost TuningSAFE
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Overview
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
4f83675ca38aOBSERVED · 2026-10-08Host compatibility
What the documentation claims. We have not run a compatibility test.
| Host | Status | Notes |
|---|---|---|
| claude-code | mentioned |
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
--- name: gamma-cost-tuning description: 'Optimize Gamma usage costs and manage API spending. Use when reducing API costs, implementing usage quotas, or planning for scale with budget constraints. Trigger with phrases like "gamma cost", "gamma billing", "gamma budget", "gamma expensive", "gamma pricing". ' allowed-tools: Read, Write, Edit version: 1.13.0 license: MIT author: Jeremy Longshore <[email protected]> tags: - saas - gamma - api - scaling - cost-optimization compatibility: Designed for Claude Code --- # Gamma Cost Tuning ## Output Record the measurement window, aggregate usage, approved plan/workflow decision, owner, verification date, and rollback state. Do not include private presentation content, billing credentials, or user data. ## Examples Use a fictional staging workspace to compare aggregate usage after an approved asset or workflow change, then promote only when the owner accepts the evidence and publishing/privacy behavior is unchanged. ## Overview Optimize Gamma API usage to minimize credit consumption. Gamma uses a credit-based billing system where costs are driven by image generation model tier and content complexity. API access requires Pro or higher subscription. ## Prerequisites - Active Gamma Pro/Ultra/Teams/Business subscription - Understanding of credit system - Completed `gamma-install-auth` setup ## Gamma Credit System ### Image Model Tiers | Tier | Credits per Image | Quality Level | |------|-------------------|---------------| | Standard | 2-15 | Good for internal/draft presentations | | Advanced | 20-33 | Higher quality, more detail | | Premium | 34-75 | Best quality images | | Ultra | 30-125 | Highest fidelity, photorealistic | Card text generation also costs credits based on the AI model used. ### Plan Comparison | Feature | Pro | Ultra | Teams | Business | |---------|-----|-------|-------|----------| | Monthly credits | Included | More credits | Team pool | Custom | | API access | Yes | Yes | Yes | Yes | | Max cards | Standard | Up to 75 | Standard | Custom | | Ad-hoc credit purchase | Yes | Yes | Yes | Yes | | Auto-recharge | Yes | Yes | Yes | Yes | ## Instructions ### Step 1: Track Credit Usage ```typescript // src/gamma/cost-tracker.ts interface UsageEntry { generationId: string; creditsUsed: number; outputFormat: string; timestamp: Date; } class CreditTracker { private usage: UsageEntry[] = []; record(entry: UsageEntry) { this.usage.push(entry); } getDaily(): { total: number; count: number; avg: number } { const today = new Date().toDateString(); const todayUsage = this.usage.filter( (u) => u.timestamp.toDateString() === today ); const total = todayUsage.reduce((sum, u) => sum + u.creditsUsed, 0); return { total, count: todayUsage.length, avg: todayUsage.length > 0 ? Math.round(total / todayUsage.length) : 0, }; } getMonthly(): { total: number; count: number } { const thisMonth = new Date().getMonth(); const monthUsage = this.usage.filter( (u) => u.timestamp.getMonth() === thisMonth ); return { total: monthUsage.reduce((sum, u) => sum + u.creditsUsed, 0), count: monthUsage.length, }; } } // Track after each generation const tracker = new CreditTracker(); async function generateTracked(gamma: GammaClient, request: GenerateRequest) { const { generationId } = await gamma.generate(request); const result = await pollUntilDone(gamma, generationId); tracker.record({ generationId, creditsUsed: result.creditsUsed ?? 0, outputFormat: request.outputFormat ?? "presentation", timestamp: new Date(), }); return result; } ``` ### Step 2: Optimize Image Costs The biggest cost driver is image generation tier. Reduce costs by: ```typescript // EXPENSIVE: default image settings (may use Advanced/Premium tier) await gamma.generate({ content: "Company quarterly review", outputFormat: "presentation", // No imageOptions = AI chooses model tier }); // CHEAPER: explicitly use standard tier when quality isn't critical await gamma.generate({ content: "Company quarterly review", outputFormat: "presentation", imageOptions: { style: "simple flat illustration", // Simpler styles use fewer credits }, }); // CHEAPEST: text-focused, minimal images await gamma.generate({ content: "Company quarterly review", outputFormat: "document", // Documents typically use fewer images textAmount: "detailed", // Focus on text, not visuals }); ``` ### Step 3: Use Templates to Reduce Regeneration ```typescript // WASTEFUL: regenerating entire presentations for minor content changes for (const client of clients) { await gamma.generate({ content: `Proposal for ${client.name}: ${fullProposalText}`, outputFormat: "presentation", }); // Each generation costs full credits } // EFFICIENT: use templates for repeated structures // Create a one-page template gamma in the app // Then generate variations with targeted prompts for (const client of clients) { await gamma.generateFromTemplate({ gammaId: "template_proposal_id", prompt: `Customize for ${client.name}. Focus on ${client.industry}.`, exportAs: "pdf", }); // Template generations can be more cost-effective } ``` ### Step 4: Budget Alerts ```typescript // src/gamma/budget.ts const MONTHLY_BUDGET = 5000; // credits const ALERT_THRESHOLDS = [0.5, 0.75, 0.9, 1.0]; // 50%, 75%, 90%, 100% async function checkBudget(tracker: CreditTracker) { const { total } = tracker.getMonthly(); const percentUsed = total / MONTHLY_BUDGET; for (const threshold of ALERT_THRESHOLDS) { if (percentUsed >= threshold) { await sendAlert( `Gamma budget ${(threshold * 100)}% used: ${total}/${MONTHLY_BUDGET} credits` ); } } // Hard stop at 100% if (percentUsed >= 1.0) { throw new Error(`Monthly Gamma budget exceeded: ${total}/${MONTHLY_BUDGET} credits`); } } ``` ### Step 5: Caching to Avo
Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | PASS |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
4f83675ca38afull audit observations/trust-audit/skill/jeremylongshore__gamma-cost-tuning.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-08 | 4f83675ca38a | SAFE | B | 89 | first audit |
Questions
What does the Gamma Cost Tuning skill do?
Model-agnostic agent-skills platform with a harness-free canonical layer, verified adapters, and the ccpi package manager. Explore at tonsofskills.com.
Is Gamma Cost Tuning safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can Gamma Cost Tuning access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
Which assistants does Gamma Cost Tuning work with?
Its documentation mentions claude-code. That is what the text claims, not a compatibility test we ran.
How current is this page?
The grade is for one exact copy of the source (4f83675ca38a), read on 2026-10-08. The repository is watched, and a new audit runs when it changes — this is the first audit.