Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    AgentSpend: Apify, MCP & LLM FinOps Circuit-Breaker

    1

    Prevent 5x surprise Apify, proxy and LLM bills by auditing true cost-per-valid-result and tripping circuit-breakers on >20% bot-block loops.

    $29

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    AgentSpend: Apify, MCP & LLM FinOps Circuit-Breaker

    AgentSpend: Apify, MCP & LLM FinOps Circuit-Breaker

    Example session with this skill installed

    Audit our day-10 Apify Actor runs and LLM enrichment logs against monthly_budgets.json and flag any retry or proxy burn spirals.

    • Read your context and instructions
    • Compiled the agentspend

    OPEN circuit-breaker alert on ecom-price-actor (38.4% anti-bot block rate burning 18.5 GB / $148 residential proxy for 4.8x projected monthly overrun), 84% mid-cycle warning on llm-enricher, true cost-per-valid-item ledger, and CIRCUIT_BREAKER_POLICY.json with capped maxItems.

    true cost-per-valid-item ledger, and CIRCUIT_BREAKER_POLICY.json

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Autonomous scraping and agent workflows silently burn residential proxy gigabytes and uncached LLM tokens when anti-bot blocks trigger retry loops—turning a $150 expected monthly budget into a 5x surprise invoice.

    What your AI assistant does with it

    • True cost-per-valid-result: Unpacks Compute Units, $8/GB residential proxies, pay-per-event fees and LLM tokens divided by valid items returned.
    • >20% bot-block spiral detector: Trips an immediate alert when failed or blocked requests cross 20% while residential proxy or token burn continues.
    • Mid-cycle 5x burn projection: Projects end-of-month spend velocity against workflow and global budgets with 80% warning and 100% open-circuit gates.
    • Ready-to-apply maxItems caps: Generates CIRCUIT_BREAKER_POLICY.json with safe maxItems, maxTotalChargeUsd and timeout ceilings per Actor.

    How a run works

    1. Export Apify Usage/run stats and MCP or LLM usage logs alongside monthly caps
    2. Run agentspend_guard.py to compute unit economics and error/block ratios
    3. Inspect tripped circuit-breakers, zero-yield runs and >20% unit-cost drift
    4. Approve gated Actor stops and apply the generated maxItems & charge caps

    Built for: AI automation agency owners, Scraping & data ops engineers, Platform & FinOps leads.

    Included: Skill instructions with live-run abort STOP AND VERIFY gates, agentspend_guard.py: Apify/MCP/LLM unit-cost auditor & circuit-breaker engine, CIRCUIT_BREAKER_POLICY.json, Slack alert cards, Excel XML & HTML dashboard, Apify CU/proxy/PPE & MCP FinOps billing mechanics playbook, Proxy retry-spiral evaluation case.

    Part of the B2B Operations & FinOps suite by Production AI Agents. Every calculation runs locally in standard-library Python (Decimal math, zero hidden network calls) and pauses at a human STOP AND VERIFY gate before any external write or sign-off.

    Analyzes usage exports and generates gated abort/cap policies locally; never terminates live production Actors without explicit confirmation. Always review generated calculations, drafts and SQL/API payloads before executing external actions.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed today

    Requires

    Python 3.10+
    Python

    Permissions

    File Scopes

    Apify
    run/usage
    exports
    MCP/LLM
    token
    logs
    and
    budget
    policy
    JSON
    (read);
    spend
    ledger
    CSV
    Excel
    XML
    HTML
    report
    (write)

    agentspend_guard.py audits usage exports offline with zero network calls and scrubs apify_api_* / sk-* tokens. Live Actor stops require explicit human confirmation.

    Listedtoday

    What's inside

    Frequently Asked Questions

    When a target site tightens anti-bot protection, retry loops keep consuming residential proxy gigabytes and LLM tokens while returning zero valid items. AgentSpend tracks valid-item yield and error rates together.

    No. It emits exact CIRCUIT_BREAKER_POLICY.json caps and abort commands, then pauses at a STOP AND VERIFY gate so an operator confirms before terminating live runs.

    Yes. Alongside Apify CU, proxy GB and pay-per-event charges, it audits input/output/cached LLM token spend per workflow and flags uncached prompt overhead.

    New in LLM Ops