More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Stock Photo JPEG Checker Preflight

    1

    Review JPEG dimensions, embedded color-profile evidence and metadata links before a stock-photo upload. Python 3.11+ required.

    $5

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Stock Photo JPEG Checker Preflight

    Example session with this skill installed

    Using the installed 0.1.10 Skill and CLI, check my selected private local folder of JPEGs and metadata CSV. Return the full JSON receipt; leave inputs unchanged and do not upload. Seller demonstration: synthetic problem and separately prepared correct samples, recorded on Windows/Python 3.12.14.

    • Read your context and instructions
    • Compiled the stock photo jpeg

    Recorded synthetic Windows CLI results

    The exact installed 0.1.10 buyer package was run on Windows/Python 3.12.14 with synthetic local files. These are complete JSON results, not a claimed hosted-agent or MCP test. The correct input was prepared separately; this checker did not repair the problem input.

    Problem sample

    {
      "findings": [
        {
          "code": "SRGB_PROFILE_NOT_EMBEDDED",
          "file": "sample.jpg"
        }
      ],
      "metadata_csv": {
        "bytes": 74,
        "file": "metadata.csv",
        "rows": 1
      },
      "photos": [
        {
          "bytes": 63129,
          "color_profile": "unknown",
          "file": "sample.jpg",
          "height": 2000,
          "width": 2000
        }
      ],
      "scope": "technical subset only; no Adobe approval or visual/legal/AI-origin decision",
      "status": "REVIEW"
    }
    

    Separate correct sample

    {
      "findings": [],
      "metadata_csv": {
        "bytes": 74,
        "file": "metadata.csv",
        "rows": 1
      },
      "photos": [
        {
          "bytes": 63735,
          "color_profile": "sRGB embedded",
          "file": "sample.jpg",
          "height": 2000,
          "width": 2000
        }
      ],
      "scope": "technical subset only; no Adobe approval or visual/legal/AI-origin decision",
      "status": "PASS_CHECKED_RULES"
    }
    

    Next steps

    Inspect every finding, correct only a separate working copy, then repeat the documented local check. Use the platform native review separately; PASS_CHECKED_RULES is not platform acceptance.

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

    About this skill

    Review JPEG dimensions, embedded color-profile evidence and metadata links before a stock-photo upload.

    DOWNLOAD TYPE A local Python CLI with companion Agent Skill instructions and an optional local MCP adapter. Python 3.11+, trusted pip, setuptools>=68 and wheel, plus Pillow 11.3.0 installed separately. Selected files must be in a private local folder. Resource-limited parsing requires a supported Windows or Linux host; see the README.

    WHO IT IS FOR Stock contributors and technical assistants who already work with Python and want a repeatable local technical review.

    CHECKS INCLUDED

    • JPEG size and dimensions
    • Embedded sRGB evidence
    • Metadata filename linkage

    REAL LOCAL DEMO The synthetic JPEG has no embedded ICC profile, so the tool asks for review rather than guessing sRGB. The measured problem sample returned REVIEW; the separate correct sample returned PASS_CHECKED_RULES. Synthetic fixtures were run through the exact installed 0.1.10 buyer package on Windows/Python 3.12.14. These are CLI results, not a hosted-agent compatibility test.

    WHAT YOU RECEIVE The current 0.1.10 ZIP, local CLI source, Agent Skill instructions, optional MCP adapter, usage guide and packaged examples where supplied. One-time current download; no subscription is needed by the CLI.

    LIMITS Up to 100 top-level JPEGs; 4 to 100 MP; conservative 45,000,000-byte photo cap. Optional UTF-8 CSV: 1,000,000 bytes and 5,000 rows. Color profiles other than the pinned canonical sRGB profile need manual review, including valid sRGB variants. Not checked: Visual quality, rights, releases, keyword relevance, AI labeling and Adobe acceptance.

    PRIVACY AND WORKFLOW The checker processes selected local files and prints JSON. It does not edit, merge, upload or import them and does not sign in to your platform. Photo filenames and technical metadata appear in the report; pixels and raw metadata titles/keywords do not.

    ALTERNATIVES Adobe's contributor portal and your own image editor can perform other necessary checks. This product adds a repeatable local JSON workflow; it does not replace those free tools.

    Use the destination platform's native workflow after correcting or reviewing findings. A local PASS is not platform approval. No affiliation or endorsement is implied.

    Up to 100 top-level JPEGs; 4 to 100 MP; conservative 45,000,000-byte photo cap. Optional UTF-8 CSV: 1,000,000 bytes and 5,000 rows. Color profiles other than the pinned canonical sRGB profile need manual review, including valid sRGB variants.
    Not checked: Visual quality, rights, releases, keyword relevance, AI labeling and Adobe acceptance.

    Parser execution: a fixed local Python child is given a 1 GiB committed-memory/address-space limit, 20 CPU seconds and a 25-second parent wall-clock deadline, with 128 KiB caps per output stream. Limits are installed before untrusted parsing on Windows Job Objects or Linux resource limits; unsupported hosts fail closed. This is resource containment, not a privilege, filesystem or network sandbox. The tool does not protect against arbitrary code execution in a vulnerable installed decoder. Install trusted runtimes and inspect only files from a source you trust. A selected local private folder is required.

    A folder and its ancestors remain bound through enumeration and reads; symlinks/reparse points, mapped network paths and hardlinked file inputs are refused. Linux reads use the held directory descriptor. Batch parser admission stops after 120 wall seconds; cleanup can add a short bounded wait. Resource failures produce REVIEW/ERROR, never PASS.

    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

    Stock Photo Preflight MCP
    Python

    Permissions

    Terminal / Shell
    Read Files
    Write Files

    Local buyer-selected files only. Terminal executes the included Python CLI. Write Files is for installation and saving receipts, never modifying inputs. No account, remote service, browser or network access is needed for the checks.

    Listed1 month ago

    What's inside

    Frequently Asked Questions

    No. It is a downloadable Python CLI plus Agent Skill. You need the listed Python prerequisites.

    No. It returns a selected-rule review report. You decide any edits and carry out the native platform workflow.

    No. CLI checks are independent of MCP and AI. Optional agent/MCP use requires your own compatible host and separate dependencies.

    Only that this tool found no issue in the documented subset. It is not certification or platform acceptance.

    Synthetic local files were checked with the exact current buyer ZIP on Windows/Python 3.12.14. These are CLI tests; optional hosted agent and MCP compatibility were not retested.

    New in Workflow Automation