agentleFS
0xatd0xatd/cheaptokens-skills/skills/venice-billing/SKILL.md

venice-billing

Skill2 starsChanged 5 months agoMITView on GitHub

Venice billing and usage analytics - GET /billing/balance, GET /billing/usage (paginated per-request ledger, JSON or CSV), and GET /billing/usage-analytics (aggregated by date/model/key). Covers the DIEM/USD/BUNDLED_CREDITS consumption priority and building dashboards. (Beta)

  • Reads credentials

1 other project uses a version of this, 1 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All require Bearer auth (not x402 — for wallet balances, use venice-x402).

    line 16

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Venice Billing

Three read-only endpoints for account-level billing and analytics. All are under a Beta tag — schema/behavior may change.

EndpointPurpose
GET /billing/balanceCurrent canConsume flag, remaining DIEM & USD, epoch allocation.
GET /billing/usagePaginated per-request ledger. JSON or CSV.
GET /billing/usage-analyticsAggregated breakdowns: by date, model, API key.

All require Bearer auth (not x402 — for wallet balances, use venice-x402). GET /billing/balance and GET /billing/usage require an ADMIN key — an INFERENCE key gets 401. GET /billing/usage-analytics works on any authenticated key (scoped to the account behind the key).

Currency / priority

Venice debits from, in order:

  1. DIEM — staked credits (reset per epoch).
  2. BUNDLED_CREDITS — included in some Pro plans.
  3. USD — prepaid fiat balance.
  4. (VCU) — deprecated legacy DIEM.

consumptionCurrency on /billing/balance reports the current currency being consumed.

GET /billing/balance

curl https://api.venice.ai/api/v1/billing/balance \
  -H "Authorization: Bearer $VENICE_API_KEY"
{
  "canConsume": true,
  "consumptionCurrency": "DIEM",
  "balances": { "diem": 90.5, "usd": 25 },
  "diemEpochAllocation": 100
}
  • canConsume: false means both DIEM and USD buckets are empty on this endpoint — canConsume here is hasPositiveDiemBalance || usdBalance > 0 and does not factor in bundled credits (which are consulted during the actual request in getConsumableBalanceForRequest).
  • consumptionCurrency is "DIEM", "USD", or null (when neither applies).
  • balances.diem is null if not staking.
  • diemEpochAllocation is the ceiling for the current epoch — balances.diem / diemEpochAllocation = remaining fraction.

GET /billing/usage

Paginated per-request ledger.

curl "https://api.venice.ai/api/v1/billing/usage?limit=200&page=1&sortOrder=desc&currency=USD&startDate=2026-04-01T00:00:00Z&endDate=2026-04-21T23:59:59Z" \
  -H "Authorization: Bearer $VENICE_API_KEY" \
  -H "Accept: application/json"

Query parameters

ParamNotes
currencyUSD / VCU / DIEM / BUNDLED_CREDITS.
startDate / endDateISO 8601 datetime.
limit1–500. Default 200.
pageDefault 1.
sortOrderasc / desc on createdAt. Default desc.

Accept header

  • application/json (default) — paginated JSON.
  • text/csv — downloads billing-usage.csv (sets Content-Disposition).

Response (JSON)

{
  "warningMessage": "DIEM (formerly VCU) has been renamed...",
  "data": [
    {
      "timestamp": "2026-04-20T12:34:56Z",
      "sku": "zai-org-glm-5-1-llm-output-mtoken",
      "units": 0.000227,
      "pricePerUnitUsd": 2.8,
      "amount": -0.06356,
      "currency": "DIEM",
      "notes": "API Inference",
      "inferenceDetails": {
        "requestId": "chatcmpl-...",
        "promptTokens": 339,
        "completionTokens": 227,
        "inferenceExecutionTime": 2964
      }
    }
  ],
  "pagination": { "limit": 200, "page": 1, "total": 1000, "totalPages": 5 }
}

Response headers: x-pagination-{limit,page,total,total-pages}.

Fields

  • sku — billing line item (model + unit type + format).
  • units — for LLMs, millions of tokens (e.g. 0.000227 = 227 tokens).
  • pricePerUnitUsd — rate; for DIEM, DIEM ≈ USD so this doubles as reference.
  • amount — negative for debit.
  • inferenceDetails — present for inference SKUs; requestId is the id returned on the original /chat/completions response.

GET /billing/usage-analytics

Aggregated summary for dashboards. Cached 10 minutes.

curl "https://api.venice.ai/api/v1/billing/usage-analytics?lookback=7d" \
  -H "Authorization: Bearer $VENICE_API_KEY"

Query parameters (choose one approach)

  • lookback=Nd — 7d, 30d, up to 90d. Default 7d.
  • OR startDate=YYYY-MM-DD + endDate=YYYY-MM-DD — both required if either is given.

Response (selected keys)

{
  "lookback": "7d",
  "byDate": [{ "date": "2026-04-20", "USD": 0.5, "DIEM": 10.25 }, ...],
  "byModel": [
    {
      "modelName": "GLM 5.1",
      "unitType": "tokens",
      "modelType": "LLM",
      "totalUsd": 0.4,
      "totalDiem": 12.5,
      "totalUnits": 50000,
      "breakdown": [
        { "type": "Output", "usd": 0.3, "diem": 10, "units": 35000 },
        { "type": "Input",  "usd": 0.1, "diem": 2.5, "units": 15000 }
      ]
    }
  ],
  "byModelDaily": [
    { "date": 1705276800000, "GLM 5.1": 5.5, "Claude Opus 4.7": 3.2 }
  ],
  "byModelDailyUsd": [...],
  "topModels": ["GLM 5.1", "Claude Opus 4.7"],
  "byKey": [
    { "apiKeyId": "key_abc123", "description": "Production Key",
      "totalUsd": 0.8, "totalDiem": 15, "totalUnits": 75000 },
    { "apiKeyId": null, "description": "Web App",
      "totalUsd": 0, "totalDiem": 4, "totalUnits": 25000 }
  ],
  "byKeyDaily": [...],
  "byKeyDailyUsd": [...],
  "topKeyNames": [...]
}
  • byDate / byModelDaily / byKeyDaily are pre-shaped for time-series charts.
  • topModels / topKeyNames give top-8 names for legend rendering.
  • apiKeyId: null in byKey means the usage originated from Venice's web app.

Recipes

Abort before calling inference if balance is empty

const { canConsume } = await fetch(`${base}/billing/balance`, { headers }).then(r => r.json())
if (!canConsume) throw new Error('Venice balance exhausted — top up before continuing')

Monthly CSV export

curl "https://api.venice.ai/api/v1/billing/usage?startDate=2026-04-01T00:00:00Z&endDate=2026-04-30T23:59:59Z&limit=500" \
  -H "Authorization: Bearer $VENICE_API_KEY" \
  -H "Accept: text/csv" \
  -o billing-april.csv

Paginate via page=1,2,3,... until page > totalPages.

Top-models chart

const a = await fetch(`${base}/billing/usage-analytics?lookback=30d`, { headers }).then(r => r.json())
// chart(a.byModelDaily, { series: a.topModels, xField: 'date' })

Errors

CodeMeaning
400Bad params (startDate without endDate, calendar range > 90 days). lookback=100d is silently clamped to 90 days rather than rejected.
401Auth failed, or INFERENCE key used on /billing/balance or /billing/usage (ADMIN required).
500Internal error.
504Analytics query timed out — shorten lookback or date range.

Gotchas

  • This is Beta — field names may shift. Validate against swagger.yaml periodically.
  • currency values include legacy VCU — use DIEM instead in new code.
  • inferenceDetails is null for non-inference SKUs (e.g. subscription charges).
  • The analytics endpoint is cached 10 min — sudden spikes lag in the dashboard by that window.
  • byModelDaily.date is a Unix milliseconds integer; byDate.date is a YYYY-MM-DD string. Don't mix them.
  • Usage entries from the Venice web app have apiKeyId: null — don't drop them when reconciling.
  • For x402 (wallet) balance, don't use this endpoint — use GET /x402/balance/{walletAddress}.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.