Provides search over Openverse's index for openly licensed images and audio, with filters, random picks, and ready-to-paste attribution on every result.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@free-image-library Find a CC-licensed photo of a sunset for my blog header, with attribution."
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Free images for any project: millions of openly-licensed, royalty-free Creative Commons images and audio (Flickr, Wikimedia, museums) — every result with ready-to-paste attribution. Zero API keys, zero configuration.
The visual sibling of the Free Music Library for the same content-creator audience: the agent finds the image AND the credit line in one call, so the attribution never gets lost between copy and paste.
Models suggest random stock images (license unknown, credit never written). This returns license-safe images with the credit the user MUST paste — built into the same result.
License intelligence, not just labels: every row carries a credit_note
stating the actual obligation — CC BY-NC rows are loudly flagged
non-commercial-only before your client ships them into a paid campaign.
Serendipity: image_roulette picks from deep in the result set, not the
first hit everyone has seen.
Built on Openverse — keyless (anonymous limits:
20 requests/min, 200/day, verified; a free registered key raises them via
OPENVERSE_API_KEY). A 5-minute TTL cache makes repeat searches free.
Related MCP server: webfetch
Tool | What it does |
| Free CC image search with filters (license, aspect_ratio, size, extension) + attribution on every row |
| One random license-safe pick, honest about how it was chosen |
| Free CC music / sound effects / field recordings from Openverse's audio index |
| Updatable playbooks: CC license briefs in plain language, error recovery |
Plus prompts: hero-image, license-safe-images.
# 1-Line Universal Installer (auto-configures Claude Desktop, Cursor, Claude Code, VS Code, ...)
curl -fsSL "https://free-image-library.builditwithai.xyz/install" | bash
# Or run directly via your preferred runtime:
uvx free-image-library
npx -y free-image-libraryUser: I need a hero image for my climate blog post
find_images(query="melting glacier", aspect_ratio="wide", count=3,
intent="hero image for a climate blog post")
→ images: [{
title: "Glacier calving", creator: "…", license: "CC BY 2.0",
image_url: "https://…", width: 2048, height: 1152,
attribution: "\"Glacier calving\" image by … (flickr), licensed CC BY 2.0 (…)",
credit_note: "Credit REQUIRED — paste the attribution verbatim." }]Anonymous usage telemetry (no PII, no queries, no paths) via the fleet
standard (schema v2, dual-endpoint fallback). Opt out any time:
FREE_IMAGE_LIBRARY_TELEMETRY=false or DO_NOT_TRACK=1.
uv venv && uv pip install -e ".[dev]"
DO_NOT_TRACK=1 .venv/bin/python -m pytest tests/ -q # unit + live + e2eLive tests hit the real Openverse API; they skip themselves when offline or rate-limited.
MIT
Search CC music, sound effects and field recordings from Openverse's audio index
| Name | Required | Description | Default |
|---|---|---|---|
| count | No | ||
| query | Yes | ||
| intent | No | ||
| license | No | cc |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already communicate that this is read-only, idempotent, and open-world, so the safe search behavior is clear. The description adds useful context about the content types and source, but it does not disclose result shape, pagination, or license handling beyond what the schema and annotations already provide.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single front-loaded sentence with no filler. Every word contributes to explaining the resource and scope of the search.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool's domain and purpose are clear, and the required query parameter lowers invocation risk. However, there is no explanation of what the response contains, what format results take, or what the intent parameter means, leaving meaningful gaps for an agent.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, and the description provides no guidance for query, count, intent, or license. The agent must infer meaning from parameter names and defaults alone, which is a significant gap for a 4-parameter search tool.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description is specific and actionable: it says the tool searches CC music, sound effects, and field recordings from Openverse's audio index. This clearly distinguishes it from the sibling find_images tool and other non-search tools.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description makes it clear that this is the audio search tool, so an agent can infer when to use it. However, it gives no explicit guidance about when not to use it or when an alternative like find_images would be preferable; that exclusion is only implied by the sibling names.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Search millions of CC-licensed images (Flickr, Wikimedia, museums…); every result carries a ready-to-paste attribution
| Name | Required | Description | Default |
|---|---|---|---|
| size | No | ||
| count | No | ||
| query | Yes | ||
| intent | No | ||
| license | No | cc | |
| extension | No | ||
| aspect_ratio | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already cover read-only, idempotent, and open-world behavior. The description adds an important behavioral detail not present in the annotations: every result carries a ready-to-paste attribution. This helps the agent know the output format and fit for attribution-related workflows beyond what the structured annotations disclose.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence that states the core functionality and a key output characteristic without repetition or fluff. It is front-loaded with the main action and includes useful source context in parentheses...
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the 7 parameters with 0% schema description coverage and no output schema, the description carries a heavy burden and leaves too many open questions. It does cover the basic licensing and attribution aspects but not the meaning/ge of the optional parameters, expected return shape, or any pagination/limit behavior. This is barely adequate guidance for a low-stakes read-only tool but well short of what is needed for nuanced parameterized searches.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
None of the seven parameters are documented, and schema description coverage is 0%. The description only hints at 'query' via 'Search' and at the 'license' field via 'CC-licensed', while the many other parameters (size, count, intent, extension, aspect_ratio) receive no explanation. Care must be given to the defaults and beyond, without which the agent can only guess at intent.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('Search'), the resource ('CC-licensed images'), and adds source examples (Flickr, Wikimedia, museums) plus an output feature (ready-to-paste attribution). This lets an agent distinguish the tool from siblings like find_audio and image_roulette without needing to open the schema.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies the tool is appropriate when looking for openly-licensed images with usable attribution, but it does not explicitly state when to use it versus alternatives such as image_roulette or find_audio. There is no mention of exclusions, edge cases, or situations where another tool would be preferable, so the guidance remains implicit rather than actionable.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Serendipity: a random openly-licensed image matching a topic — one pick, honest about how it was chosen
| Name | Required | Description | Default |
|---|---|---|---|
| seed | No | ||
| query | No | ||
| license | No | cc |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
The description adds non-obvious behavioral signals beyond annotations: the tool returns a single result, the selection is random/non-deterministic, and it is 'honest about how it was chosen.' This complements the readOnlyHint and aligns with idempotentHint=false; no contradiction is present.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
One deliberately hand-crafted sentence fits the whole essence of the tool: randomness, scope, licensing, and the "one" pick. There is no filler, and the key behavioral distinction is front-loaded.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple read-only tool the description is relatively complete, but it lacks an output schema and does not describe what shape the result takes (URL, metadata, etc.). The seed parameter remains a mystery, so a fully correct call from a cold start is still uncertain.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema description coverage, the description needs to carry the burden of explaining all parameters. It loosely maps 'matching a topic' to query and 'openly-licensed' to license, but it never explains the seed parameter, allowed license values, or the relationship between seed and randomness.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly identifies a distinct operation: returning a single random, openly-licensed image matching a topic. 'One pick' and 'random' distinguish it from the adjacent find_images tool, so there is little ambiguity about what this tool does.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The use case is implied by 'serendipity' and 'one pick' — use when you want a chance discovery rather than a broad search. However, it does not explicitly name when to use this vs. find_images or give any exclusions/preconditions, so the agent must infer this distinction.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Fetch the full content of one skill by name (from skills_list)
| Name | Required | Description | Default |
|---|---|---|---|
| name | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, openWorldHint, and idempotentHint, so the safety and repeatability profile is known. The description adds minimal behavioral context beyond purpose—it says 'full content' (return size) and references the source of the name. It does not contradict annotations and adds small value without redundancy.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single sentence with no filler. The core action is front-loaded ('Fetch the full content of one skill'), followed by the key constraint ('by name') and the source. Every word earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
This is a simple, read-only tool with one parameter and no output schema. The description tells the agent what it returns ('full content') and where the name comes from. Combined with annotations, the agent has everything needed to call it correctly. Nothing essential is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, so the description must compensate. It does by specifying that the 'name' parameter is the skill name taken from skills_list. This adds meaning beyond the bare schema property, telling the agent how to obtain a valid value. It does not define format, but the source is sufficient for a single string parameter.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description states a specific verb ('Fetch'), a resource ('full content of one skill'), and the key parameter ('by name'). It also references the source of the name ('from skills_list'), which distinguishes it from sibling tools like skills_list that list all skills. This is unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context: you need a skill name, and that name comes from skills_list. It does not explicitly name alternatives or when-not-to-use, but the reference to skills_list gives clear guidance on where names originate. This provides enough context without being explicit about exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List available skills (guidance playbooks) for using this server well — read one with skill_read
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true and idempotentHint=true, covering the safety profile. The description adds meaningful context by defining what skills are (guidance playbooks) and their purpose (using the server well). It does not contradict annotations and provides useful domain context beyond the structured fields.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
A single sentence that front-loads the core action ('List available skills'), immediately defines the resource, and ends with an actionable pointer to the sibling tool. Every word earns its place; there is no fluff.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple, parameterless, read-only listing tool, the description is complete. It tells the agent what the tool returns (a list of skills), what those skills are, and how to proceed (skill_read). No output schema exists, but the simplicity makes that acceptable. The annotations cover safety, so nothing is missing.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters, so there is nothing to document. Per the rubric, a baseline of 4 applies for 0 params. The description adds no parameter information because none exists, and the schema correctly shows an empty properties object.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('List') and a resource ('available skills'), clarifies they are 'guidance playbooks', and directly distinguishes from the sibling tool skill_read by pointing to it as the next step. This is a precise, unambiguous purpose statement.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly tells the agent when to use this tool (to see the list of skills) and when to use the alternative skill_read ('read one with skill_read'), providing clear routing between the two related tools. This matches the standard of the get_calls example.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0find_audiofind_imagesimage_rouletteskill_readskills_listScored across 5 tools
find_images and image_roulette both involve images, but one is a general search and one is explicitly a random single pick, so they are unlikely to be conflated. find_audio is clearly a separate media type, and skills_list/skill_read form an obvious complementary pair.
The find_images and find_audio names establish a find_<resource> pattern, while image_roulette breaks that pattern with a noun-style name. skills_list and skill_read are internally consistent but use a different <resource>_<verb> ordering, creating a minor but noticeable inconsistency.
Five tools is a well-scoped count for this server's purpose. Each tool has a distinct role, and the skills-related pair supports agent usage without bloating the surface.
The image and audio search workflows are well covered, including a serendipity mode and attribution in results. The only minor gap is the lack of a dedicated detail/fetch tool for a single media item by identifier, though this may not be necessary for an external-library search use case.
Semantic search over free-to-use stock photos from 9 libraries: by words, image, or similar.
Personal asset search engine: everything you make or upload is searchable by what's inside it.
Search 14.5M Smithsonian Open Access objects, get CC0 images, find cross-collection connections.
