Skip to content
ToolCargo

A practical MCP workflow

Research NASA image records with AI and MCP

Use ToolCargo to search NASA’s image library, inspect a selected record by its exact ID and prepare an illustration shortlist with source dates and recorded credits. ToolCargo projects the retrieved metadata; NASA has not reviewed the resulting AI output. Review original reuse guidance before selecting an image.

Built for: Editors, educators and research agents preparing space or mission illustration shortlists.

What to connect

Create a ToolCargo account and use OAuth or an API key with a supported MCP client. Hosted connectors share your plan’s call quota; connect each required MCP endpoint separately. Review provider permissions before starting.

Client setup and test status · Current plans and limits

Run the workflow

  1. 1. Define the topic and source filters

    Call nasa_search_images with a human search phrase and a small limit. Optional center and year filters refine the indexed records. Year filters and dateCreated refer to the library’s source field; they do not establish an image’s capture date, upload date or event date. Search uses the provider’s text semantics, and ranking can change.

  2. 2. Inspect a selected identifier

    Copy one returned nasaId into nasa_image_record. The tool compares the returned image ID exactly ignoring ASCII case, preserves canonical source spelling and discloses trimmed input. A valid empty index response means not_indexed in this search source, not that the asset has never existed. A failed or malformed source response produces an error.

  3. 3. Keep caption, dates and credits separate

    Retain the source title, bounded description, center and recorded photographer or secondary-creator fields. Missing credits stay missing; center is not automatically the creator. A date mentioned in a caption may differ from dateCreated. Do not silently rewrite these into one authoritative capture date or infer missing authorship.

  4. 4. Compare a separate Commons candidate if useful

    Search Wikimedia Commons independently and inspect a chosen exact File: title with commons_file_metadata. Compare its recorded dimensions, creator credits and terms with the NASA library candidate. The two providers have different IDs and records; do not assume they describe the same image or share reuse permissions.

  5. 5. Prepare a source-linked shortlist for review

    Show each identifier, original library pointer, source dates, recorded credits, missing fields and fetchedAt/cacheHit. Attribute the generated synopsis to the AI product and disclose that it includes NASA source-library material without implying NASA review or endorsement. Reuse remains unassessed: inspect source guidance, third-party rights, identifiable people and brand restrictions before using any file.

A prompt to try

Use ToolCargo NASA Image Library Research to find image records for the topic I provide. Inspect selected exact nasaId records and retain canonical IDs, source dates, recorded credits, missing fields and source links. Compare separately selected Wikimedia Commons records if useful, without assuming cross-source identity or shared permissions. Label your generated synopsis as your AI product’s output and disclose that it includes NASA source-library material; do not use “according to NASA” wording or imply NASA reviewed or endorsed it. Leave reuse rights unassessed. Do not fetch image files, train models, generate images or invent capture dates, credits or licenses.

What a useful result looks like

An AI-prepared illustration shortlist: provider library → exact source ID → title → recorded dateCreated → separately quoted caption date if present → recorded credits or missing → source link → fetchedAt/cacheHit → fields to review → reuseStatus not_assessed. Keep each provider’s record and rights context separate.

Know the limits

This workflow researches indexed image metadata. It does not fetch media, follow asset manifests, verify image contents or grant reuse permission. NASA source-library disclosure is separate from authorship of AI output. Records can include third-party content, identifiable people and protected marks; public indexing does not establish public-domain status or clearance. Search pages can drift and a bounded cache introduces disclosed delay. Each successful hosted tool call, including a cache hit, uses the shared ToolCargo allowance; no NASA provider key or paid provider is needed.

Common questions

Does dateCreated establish when the photograph was taken?

No. It is a source-library field. Preserve it separately from dates in the caption and verify the original record if the historical date matters.

Are NASA library images automatically cleared for reuse?

No. Reuse remains unassessed. Review NASA’s current guidelines, recorded third-party credits, identifiable-person rights and brand restrictions for the intended use.

Is the generated shortlist a NASA-authored report?

No. The AI product prepares its own output using disclosed source-library material. NASA has not reviewed or endorsed the projection or resulting synopsis.

Can another search page be treated as the same snapshot?

No. Continue explicitly with unchanged query, filters and page size, but records and rankings can change between calls. The source collection version is not a frozen revision.

References and tool documentation

Use the provider’s documentation to check the underlying concepts, and ToolCargo’s references for the exact tools, inputs and limits.

Tool references for this workflow

Continue with the tools

Related workflows