Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    OpenAI Batch Result Reconciliation

    1

    Join submitted Batch JSONL requests with downloaded output and error JSONL by custom_id, retain input order, and create an explicit missing/failed/duplicate reconciliation and review-only retry candidate list.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    OpenAI Batch Result Reconciliation

    OpenAI Batch Result Reconciliation

    Example session with this skill installed

    Illustrative invocation for the seller's synthetic acceptance fixture: reconcile two submitted Batch JSONL custom_ids against reversed-order downloaded successful outputs and an empty error file. Preserve request order. Do not call APIs or submit retries.

    • Read your context and instructions
    • Compiled the openai batch result

    Seller demonstration — synthetic fixture

    This is the complete recorded JSON from the seller's local CLI replay of the exact accepted ZIP, version 0.2.0. It is not a live account export or a verified Agensi-hosted agent session. The invocation above is illustrative; the output below is the preserved actual seller test result.

    Complete observed output

    {
      "coverage": {
        "errors": {
          "bytes": 0,
          "reader": "jsonl",
          "records": 0,
          "source_complete": "NOT_VERIFIED"
        },
        "outputs": {
          "bytes": 288,
          "reader": "jsonl",
          "records": 2,
          "source_complete": "NOT_VERIFIED"
        },
        "requests": {
          "bytes": 264,
          "reader": "jsonl",
          "records": 2,
          "source_complete": "NOT_VERIFIED"
        }
      },
      "details": {
        "aligned_outcomes": [
          {
            "code": "HTTP_200",
            "custom_id": "A",
            "file": "outputs",
            "input_record": 1,
            "outcome": "SUCCEEDED",
            "record": 2
          },
          {
            "code": "HTTP_200",
            "custom_id": "B",
            "file": "outputs",
            "input_record": 2,
            "outcome": "SUCCEEDED",
            "record": 1
          }
        ],
        "matched_ids": 2,
        "review_only_retry_candidates": []
      },
      "findings": [],
      "input_sha256": {
        "errors": "e3b0c44298fc1c149afbf4c8996fb92427ae41e4649b934ca495991b7852b855",
        "outputs": "66fd7c7e35a47a5205a27dec478a0bc91935a58deef194ef2e6e372aa5ad5e26",
        "policy": "44136fa355b3678a1146ad16f7e8649e94fb4fc21fe77e8310c060f61caaff8a",
        "requests": "37a154605ef300929e1ca5706bbe3886c19d8e9f8131c22c5bb43e91cebfcd45"
      },
      "limitations": [
        "Missing downloaded outcomes may still arrive; live job status and billing not checked.",
        "HTTP success is not content-quality or business-success proof; no rerun is executed."
      ],
      "schema_version": "local-audit-result@1",
      "status": "CLEAR"
    }
    

    Next step

    Review the scoped findings and coverage before any manual platform action. CLEAR only covers the documented local rules; input completeness, platform acceptance and business correctness are not certified.

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

    About this skill

    Join submitted Batch JSONL requests with downloaded output and error JSONL by custom_id, retain input order, and create an explicit missing/failed/duplicate reconciliation and review-only retry candidate list.

    For: AI developers operating paid bulk inference jobs.

    Input: Native Batch input, output and error JSONL downloaded by the operator; empty output/error files allowed. No API calls or cost calculations.

    Included: original Python source, companion SKILL.md, runnable synthetic examples, policy guide, README and a file-hash manifest. Runs locally on Windows 10/11 with Python 3.11+; no third-party runtime dependencies. Use the documented source command after extraction. A Python-enabled agent is needed for Skill execution; installing the Skill alone does not include the complete source folder. Your AI assistant may have separate costs.

    Output: a JSON audit with CLEAR, REVIEW, BLOCK or INVALID_INPUT, findings, scoped details, coverage and input hashes. The audit never edits inputs, uploads data, calls a paid API or executes a platform change. CLEAR covers only the documented rules and is not platform approval or verification that your export is complete.

    Limits: No live status lookup, API-key handling, content quality scoring, charging, automated rerun or confirmation that missing results will never arrive. Input files are bounded to 1 MiB each; other structural limits are in the included README. Windows local regular files only; links, network drives, cloud placeholders and archives as audit inputs are unsupported.

    Compare before buying: Free official example Python, SDK downloads, pandas join, open-source batch wrappers. Skilled developers can self-build; a basic ID join alone would not warrant payment.

    This package adds: Fail-closed reconciliation including both error files, malformed/conflicting entries and human-reviewed missing IDs; preserves request order without issuing retries. A deliberately bounded commercial experiment, not proven demand.

    One-time download, $5. This is a bounded commercial experiment; sales and time savings are not established.

    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 1 month ago

    Requires

    Windows
    Python

    Permissions

    Terminal / Shell
    Read Files

    Reads only user-authorized local regular input files and runs the bundled Python command. Returns scoped JSON; no account access, network, browser or input changes. Source folders must remain available. Seller replay used the local CLI, not hosted agent execution.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    Join submitted Batch JSONL requests with downloaded output and error JSONL by custom_id, retain input order, and create an explicit missing/failed/duplicate reconciliation and review-only retry candidate list.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use OpenAI Batch Result Reconciliation on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in LLM Ops