agentsclimarketplace

Gemini cost tracker

Skill baronguyen001/ai-automation-skills/skills/gemini-cost-tracker

8 production-tested Claude skills: automation, Gemini cost/structured output, OSS bounty scouting, ML validation.

Install
npx -y skills add baronguyen001/ai-automation-skills --skill gemini-cost-tracker

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Wrap Gemini API calls to log token usage and USD cost per call, including input, output, cached, and thinking tokens, with per-model and per-session summaries. Use when the user asks to track gemini cost, inspect gemini token usage, or estimate repeated LLM call spend.

SKILL.md

2.0 KB, as published. Nobody here has run it

Gemini Cost Tracker

Use this skill when repeated Gemini calls need transparent token and cost accounting. It records usage metadata after each call and prints a simple per-model session summary.

When to invoke

  • User says: "track gemini cost" / "gemini token usage" / "how much is this LLM call costing"
  • Code in the conversation uses: google-genai, Gemini REST responses, or repeated model calls in a batch.

When NOT to invoke

  • The user only makes one or two ad hoc calls per day.
  • The user needs provider billing reconciliation rather than a local estimate.

Concrete example

User input:

Wrap this three-call Gemini ranking job so I can see model cost by session.

Output:

# Copy assets/cost_tracker.py into your project, then:
from cost_tracker import CostTracker

tracker = CostTracker()

# resp = client.models.generate_content(model="gemini-2.5-flash", contents=prompt)
tracker.record("gemini-2.5-flash", resp.usage_metadata)

print(tracker.summary())

Example printed output:

Calls: 3 | Total: $0.001242
  gemini-2.5-flash-lite: $0.000180
  gemini-2.5-flash: $0.001062

Pattern to apply

  1. Keep a small pricing table in USD per 1M tokens and mark it "verify before trusting."
  2. After each Gemini call, read usage_metadata from the SDK response or usageMetadata from REST.
  3. Subtract cached tokens from billable input tokens.
  4. Count output plus thinking tokens at the output-token price.
  5. Append one row per call and print a per-model summary at the end.

Reference: assets/cost_tracker.py.

Source

Distilled from production use across the author's automation projects. v0.1.0. See also: [[gemini-structured-output]].

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.