firecrawl-cost-tuning — Claude Code skill: what it does
firecrawl-cost-tuning traces where Firecrawl credits go and trims the bill through endpoint choice, format discipline, cache policy and spend limits, checked against quality.
MIT 1 installs ★ 2,935 updated 2026-10-29
npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a claude-codeWhat firecrawl-cost-tuning does
Firecrawl bills by operation and by option, so counting requests gets you nowhere. This skill maps spend onto workloads: which endpoint ran, which formats were requested, how many pages finished, whether cache kicked in. From there it points at what to cut.
The workflow opens with the current billing setup and a baseline, then sets hard bounds on crawl and batch jobs and asks for the cheapest format that still does the job. Map-first discovery comes up where listing pages costs less than fetching all of them. Freshness decides whether maxAge is worth using. Nothing goes live without a preflight budget check, alert thresholds and a canary run that weighs cost against quality.
Cloud calls need FIRECRAWL_API_KEY taken from a secret manager and sent as Authorization: Bearer; the key stays out of logs, commits and URLs. Raising pay-as-you-go, switching plans, lifting a per-key limit or trading data retention for cache savings need approval from a person. When the budget is gone, work stops — retries that bury a 402 response only hide the problem.
What it helps with
- find out why the Firecrawl bill jumped this month
- set a per-key ceiling with alerts before overage hits
- strip formats and options the job never reads
- compare crawl cost before and after a change, quality included
- check whether cache savings clash with retention rules
Example prompts for the agent
- Break down Firecrawl credit spend for
services/scraperover30dand tell me what to cut. - Why did our Firecrawl budget spike? Attribute it by endpoint and format against the baseline.
- Make this crawl cheaper and show me how page coverage changes.
How to install firecrawl-cost-tuning
1Skills CLI (Vercel)
npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a claude-code2Manually: put the files in place
~/.claude/skills/firecrawl-cost-tuning/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/firecrawl-cost-tuning/
1Skills CLI (Vercel)
npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a codex2Manually: put the files in place
.agents/skills/firecrawl-cost-tuning/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a cursor2Manually: put the files in place
.cursor/skills/firecrawl-cost-tuning/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install jeremylongshore/tons-of-skills-marketplace skills/.curated/firecrawl-cost-tuning2Manually: put the files in place
.github/skills/firecrawl-cost-tuning/SKILL.md1Skills CLI (Vercel)
npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a gemini-cli2Platform CLI
gemini skills install https://github.com/jeremylongshore/tons-of-skills-marketplace --path skills/.curated/firecrawl-cost-tuningManually: put the files in place
.devin/skills/firecrawl-cost-tuning/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about firecrawl-cost-tuning
How do I install firecrawl-cost-tuning in Claude Code?
Run npx skills add jeremylongshore/tons-of-skills-marketplace --skill firecrawl-cost-tuning -a claude-code. The skill file ends up at ~/.claude/skills/firecrawl-cost-tuning/SKILL.md.
What does firecrawl-cost-tuning do?
It attributes Firecrawl credit use by operation and workload, sets bounds for crawl and batch runs, tests whether expensive formats earn their cost, and prepares budget alerts.
Does it touch my Firecrawl account or scrape pages itself?
No. It reads code, config and tests with Read, Glob and Grep and writes only approved edits; scraping targets, rotating keys and changing account settings stay with you.
Source text
SKILL.md · MIT
--- name: firecrawl-cost-tuning description: >- Measure and reduce Firecrawl credit consumption using explicit limits, endpoint selection, cache policy, format discipline, and spend controls. Use when forecasting or correcting Firecrawl spend. Trigger with "Firecrawl costs", "reduce Firecrawl credits", or "Firecrawl budget". allowed-tools: Read,Glob,Grep,Write,Edit argument-hint: "<repository-path> <budget-window>" version: 1.12.0 license: MIT author: Jeremy Longshore <[email protected]> tags: [saas, firecrawl, cost, governance] model: inherit effort: high compatibility: "Designed for Claude Code; Firecrawl Cloud work requires network access" --- # Firecrawl Credit and Spend Control ## Overview Optimize against measured business value rather than assumed one-credit requests. Endpoint and option costs differ, modifiers can stack, and crawl or batch charges arrive as pages complete. ## Prerequisites - The target repository or integration path and the requested operator outcome. - The source authorization, data classification, and environment policy. - Current Firecrawl documentation, credentials only when needed, and an owner for approvals. ## Current Contract Use the current billing page and Credit Usage APIs as the pricing authority. Base scrape/crawl page costs, search/result charges, JSON or other format modifiers, ZDR, parse behavior, Interact minutes, and lockdown outcomes can differ. Polling status does not itself consume credits, while asynchronous page processing can make usage appear later. ## Authentication For authenticated Cloud operations, inject FIRECRAWL_API_KEY from an approved secret manager. REST requests use Authorization: Bearer with the key. Never print, commit, transmit, or place a key in a URL. Keyless access is suitable only where the current documentation explicitly allows it and the workload accepts its limits; production workflows should make identity and team ownership explicit. ## Instructions 1. Capture the current billing policy, plan, pay-as-you-go state, per-key spend limits, monthly cap, and a baseline by operation and workload class. 2. Attribute usage to an owner, environment, source policy, endpoint, requested formats, page counts, cache behavior, and success category. Never estimate solely from request count. 3. Set explicit crawl and batch bounds. Use map plus selective retrieval when discovery is cheaper than collecting every page. 4. Request only required formats and expensive options. Validate whether JSON extraction, prompt-injection checks, PDF parsing, ZDR, audio/video, or browser interaction earns its added cost. 5. Use maxAge and cache policy only when the freshness SLA permits it; use storeInCache false or ZDR when retention requirements outweigh savings. 6. Add preflight budget checks, alert thresholds, per-key controls where available, and a fail-closed response when the approved ceiling is reached. 7. Run a bounded canary, compare quality-adjusted cost with the baseline, and roll back if savings reduce completeness or violate policy. ## Tool Discipline Use Read, Glob, and Grep to inspect code, configuration, tests, and evidence. Use Write/Edit only for approved implementation or documentation changes. Do not call Firecrawl, rotate keys, change account settings, scrape a target, or deploy merely because this skill was invoked. ## Approval Boundaries Require approval before enabling or raising pay-as-you-go, changing plan, increasing a per-key spend limit, trading retention for cache savings, or reducing required quality. ## Output Return the billing-policy snapshot, workload attribution, baseline and candidate cost, quality delta, chosen controls, alerts, canary evidence, projected range, and rollback threshold. ## Error Handling - Usage lags async work: wait for job completion and billing settlement before declaring savings. - Current prices or limits cannot be verified: report a range and block irreversible plan decisions. - Budget is exhausted: stop new work; do not hide 402 responses with uncontrolled retries. ## Examples - "Why did credits spike?" attributes usage by option and completed page rather than request count. - "Make this crawl cheaper" tests scope, cache, format, and map-first changes against quality. ## Resources Read [official Firecrawl evidence](references/official-docs.md) before relying on an endpoint, SDK method, plan limit, price, retention option, or self-hosted release.
© Jeremy Longshore · MIT · jeremylongshore/tons-of-skills-marketplace