Skip to main content
Glama
io.github.pvliesdonk/image-generation-mcp

io.github.pvliesdonk/image-generation-mcp

by pvliesdonk

Image Generation MCP

CI codecov PyPI Python License Docker Docs llms.txt Template

Multi-provider image generation MCP server built on FastMCP. Generate images from Claude Desktop, Claude Code, or any MCP client using OpenAI, Google Gemini, Stable Diffusion (SD WebUI), or a zero-cost placeholder provider.

Documentation | Config wizard | PyPI | Docker

Features

  • Multi-provider: OpenAI (gpt-image-2, gpt-image-1.5, dall-e-3), Google Gemini (gemini-3.1-flash-image, gemini-3-pro-image, gemini-3.1-flash-lite-image), SD WebUI (Stable Diffusion / Forge / reForge), and a zero-cost placeholder for testing.

  • Per-model style metadata: every model carries a style_profile (strengths, prompt grammar, lifecycle); list_providers includes a top-level warnings array for deprecated models. See Model Catalog.

  • Keyword-based auto-selection: provider="auto" routes by prompt content (text/logo → OpenAI, photoreal/anime → SD WebUI, draft → placeholder).

  • CDN-style image transforms: image://{id}/view?format=webp&width=512&crop_x=... resizes / re-encodes / crops on demand without re-generating.

  • Hybrid background tasks: long-running SD generations run with task=True (poll for status); short OpenAI calls stream progress in the foreground.

  • MCP Apps gallery + viewer: interactive UI surfaces (browse generated images, edit / crop / rotate) for clients that support app: resources.

  • Production deployment: Docker (multi-arch), .deb/.rpm with hardened systemd, OIDC + bearer auth, persistent EventStore for HTTP session resumability.

Related MCP server: gemini-nano-banana-mcp

What you can do with it

With this server mounted in an MCP client, you can ask:

  • "Generate a coffee mug product photo on a worn oak table, 16:9, no text." Routes to gpt-image-1.5 for typography-aware photorealism.

  • "Create three concept-art variations of a cyberpunk alley at dusk." Composes generate_image with provider="sd_webui" and a stylised checkpoint like dreamshaperXL.

  • "Crop this image to a 1:1 square centred on the subject and resize to 512px." Uses image://{id}/view?width=512&height=512&crop_x=... resource transforms.

  • "Show me my recent generations." Browses the gallery via the image://list resource and the MCP Apps gallery viewer.

  • "Save this style as 'cyberpunk-night' so I can apply it to future requests." Uses the style library, whose markdown briefs the LLM interprets per-provider.

  • "Replace the background of my last photo with a sunset sky." Uses transform_image with the gallery image_id as a reference (image-to-image via Gemini).

Installation

From PyPI

pip install image-generation-mcp

If you add optional extras via the PROJECT-EXTRAS-START / PROJECT-EXTRAS-END sentinels in pyproject.toml, document them below:

Extra

Includes

Use when

mcp

fastmcp[tasks]>=3.0,<4

Background-task support (task=True), required for long SD generations.

openai

openai>=1.0

Enables the OpenAI provider.

google-genai

google-genai>=1.0

Enables the Gemini provider.

all

fastmcp[tasks] + openai + google-genai

Everything except SD WebUI (which is HTTP-only, no extra needed).

Example: pip install image-generation-mcp[all].

From source

git clone https://github.com/pvliesdonk/image-generation-mcp.git
cd image-generation-mcp
uv sync --all-extras --all-groups

Docker

docker pull ghcr.io/pvliesdonk/image-generation-mcp:latest

To run the newest merged code instead of the newest release, use the rolling edge tag. It is rebuilt on every merge to main and carries no version identity. See Image tags for the full tag list.

docker pull ghcr.io/pvliesdonk/image-generation-mcp:edge

A compose.yml ships at the repo root as a starting point. Copy .env.example to .env, edit, and docker compose up -d.

To attach a remote Python debugger (development only; the protocol is unauthenticated), see Remote debugging.

Linux packages (.deb / .rpm)

Download .deb or .rpm packages from the GitHub Releases page. Both install a hardened systemd unit; env configuration is sourced from /etc/image-generation-mcp/env (copy from the shipped /etc/image-generation-mcp/env.example).

Claude Desktop (.mcpb bundle)

Download the .mcpb bundle from the GitHub Releases page and double-click to install, or run:

mcpb install image-generation-mcp-<version>.mcpb

Claude Desktop prompts for required env vars via a GUI wizard, with no manual JSON editing needed.

For manual Claude Desktop configuration and setup options, see Claude Desktop deployment.

Release channels

Artifacts ship on three channels. Each row lists exactly what that channel publishes.

Channel

Version identity

Artifacts

edge (rolling)

None; the commit is the identity

Docker image :edge rebuilt on every merge to main; .mcpb bundle as the mcpb-bundle-edge workflow artifact; Claude Code plugin .zip as the plugin-zip-edge artifact; rolling unstable docs version. It leaves no git tag, GitHub release, or PyPI entry behind.

Pre-release

vX.Y.Z-rc.N, computed and reviewed in its release pull request

PyPI (as the pre-release X.Y.ZrcN); GitHub release with wheels, sdist, .deb/.rpm packages, .mcpb bundle, plugin .zip, and SBOM attached; Docker image under its immutable vX.Y.Z-rc.N tag plus the ordering-aware rolling rc tag. Skips the plugin marketplace, the MCP registry, and the docs deploy.

Stable

vX.Y.Z

Everything: PyPI, Docker (version tag plus ordering-aware latest / vX / vX.Y), .deb/.rpm, GitHub release assets (wheels, sdist, .mcpb bundle, plugin .zip, SBOM), plugin marketplace and MCP registry entries (when the release is the newest stable), versioned docs with an ordering-aware latest alias.

Pre-releases reach PyPI so that a candidate's .mcpb bundle installs: the bundle points at PyPI rather than carrying the code. Ordinary installers never see them, because a PEP 440 resolver skips pre-releases unless the requirement pins one or you pass --pre. Ask for a candidate by name with pip install image-generation-mcp==X.Y.ZrcN. PyPI spells it in the PEP 440 canonical form, while tags use SemVer. Rolling pointers are ordering-aware, so a patch release cut from an old release/X.Y branch never moves latest-style tags back to older content, and a candidate for an already-released version never moves rc. See Release process for the full model.

Quick start

image-generation-mcp serve                                # stdio transport
image-generation-mcp serve --transport http --port 8000   # streamable HTTP

For library usage (embedding the domain logic without the MCP transport), import from the image_generation_mcp package directly. See the project's domain modules under src/image_generation_mcp/ for entry points.

Server info

The server registers a built-in get_server_info tool (via fastmcp_pvl_core.register_server_info_tool) so operators can confirm the deployed version with a single MCP call. The default response carries server_name, server_version, and core_version. Servers that talk to a remote upstream wire upstream version reporting inside the DOMAIN-UPSTREAM-START / DOMAIN-UPSTREAM-END sentinel in src/image_generation_mcp/server.py; see CLAUDE.md for the wiring pattern.

Configuration

Core environment variables shared across all fastmcp-pvl-core-based services:

Variable

Default

Description

IMAGE_GENERATION_MCP_KV_STORE_URL

file:///data/state

Persistent-state backend URL shared by every pvl-core subsystem that needs state. memory:// is in-process and lost on restart; file:///path persists on one server; redis://, dynamodb:// and mongodb:// each need their matching extra. When unset, defaults to file:///data/state (the volume family Docker images mount), or to memory://; with a warning; on a host where that directory is not usable.

FASTMCP_LOG_LEVEL

INFO

Log level for FastMCP internals and app loggers (DEBUG / INFO / WARNING / ERROR / CRITICAL). The -v CLI flag overrides to DEBUG.

FASTMCP_ENABLE_RICH_LOGGING

true

Set false for plain or structured JSON log output.

Domain-specific variables go below under Domain configuration.

Authentication

Callers authenticate via a bearer token or OIDC (mutually exclusive). See the Authentication guide for setup, mapped multi-subject tokens, OIDC, and troubleshooting.

Post-scaffold checklist

After copier copy and gh repo create --push:

  1. Fill in the DOMAIN blocks (every section marked with a DOMAIN sentinel comment) in this README and in CLAUDE.md. The GENERATED-ENV-TABLE-* regions are not DOMAIN blocks; the config generator owns them and rewrites them on every run.

  2. Configure GitHub secrets (see below).

  3. Install dev + docs tooling: uv sync --all-extras --all-groups.

  4. Install pre-commit hooks: uv run pre-commit install.

  5. Run the gate locally: uv run pytest -x -q && uv run ruff check --fix . && uv run ruff format . && uv run mypy src/ tests/.

  6. Push the first commit. CI should be green.

GitHub secrets

CI workflows reference three repository secrets. Configure them via Settings → Secrets and variables → Actions or with gh secret set:

Secret

Used by

How to generate

RELEASE_TOKEN

release-prepare.yml, release.yml, release-notes.yml, copier-update.yml, renovate.yml, bootstrap.yml

Fine-grained PAT at https://github.com/settings/personal-access-tokens/new with contents: write, pull_requests: write, and administration: write (bootstrap applies the repository rulesets + auto-merge). Must belong to a repository admin: the shipped rulesets grant bypass to the admin role, and the release tag + GitHub release that knope creates after a release pull request merges rely on it (pull requests the token opens also need it so their CI runs). Scoped to this repo.

CODECOV_TOKEN

ci.yml

https://codecov.io: sign in with GitHub and add the repo. The upload token is on its settings page.

CLAUDE_CODE_OAUTH_TOKEN

claude.yml, claude-code-review.yml, release-notes.yml

Run claude setup-token locally and paste the result.

gh secret set RELEASE_TOKEN
gh secret set CODECOV_TOKEN
gh secret set CLAUDE_CODE_OAUTH_TOKEN

Dependency updates are handled by Renovate (renovate.yml), which reuses RELEASE_TOKEN. It maintains uv.lock and auto-merges patch/minor bumps once the CI Success check is green; bootstrap.yml enables auto-merge and applies the repository rulesets (.github/rulesets/) on first push. See Repository Protection for the per-branch posture and bypass model. GitHub Actions are updated in the copier template and arrive via copier update, not per-repo.

GITHUB_TOKEN is auto-provided; no action needed.

Local development

The PR gate (matches CI):

uv run pytest -x -q                                  # tests
uv run ruff check --fix . && uv run ruff format .    # lint + format
uv run mypy src/ tests/                              # type-check

Pre-commit runs a subset of the gate on each commit; see .pre-commit-config.yaml for details, or CLAUDE.md for the full Hard PR Acceptance Gates.

Troubleshooting

Moving a scaffolded project

uv sync creates .venv/bin/* scripts with absolute shebangs pointing at the venv Python. If you move the repo after scaffolding (mv /old/path /new/path), uv run pytest fails with ModuleNotFoundError: No module named 'fastmcp' because the stale shebang resolves to a different interpreter than the venv's site-packages.

Fix:

rm -rf .venv
uv sync --all-extras --all-groups

uv run python -m pytest also works as a one-shot workaround (bypasses the stale entry-script shim).

uv.lock refresh after copier update

When copier update introduces new dependencies (such as a new extra added to pyproject.toml.jinja), the CI install step runs uv sync --locked, which fails against a stale lockfile. Run uv lock locally and commit the refreshed uv.lock alongside accepting the copier-update PR.

CI installs with --locked (and the review workflow with --frozen) so no job ever rewrites uv.lock in its own workspace: a job that re-locks hides the drift it just repaired, and a dirty workspace breaks any later git checkout in the same job. Lockfile drift then shows up as a red install step with a clear message, not as a silent mutation.

Domain configuration

Domain environment variables use the IMAGE_GENERATION_MCP_ prefix:

Variable

Default

Required

Description

IMAGE_GENERATION_MCP_A1111_HOST

(none)

No

Deprecated alias for IMAGE_GENERATION_MCP_SD_WEBUI_HOST; logs a warning when used.

IMAGE_GENERATION_MCP_A1111_MODEL

(none)

No

Deprecated alias for IMAGE_GENERATION_MCP_SD_WEBUI_MODEL; logs a warning when used.

IMAGE_GENERATION_MCP_READ_ONLY

true

No

When true, write-tagged tools (image generation, transforms, uploads) are hidden from clients. Set false to enable them.

IMAGE_GENERATION_MCP_SCRATCH_DIR

~/.image-generation-mcp/images

No

Directory where generated images are saved. Created automatically on first use.

IMAGE_GENERATION_MCP_OPENAI_API_KEY

(none)

No

OpenAI API key. Enables the OpenAI provider (gpt-image-2, gpt-image-1.5, dall-e-3) when set.

IMAGE_GENERATION_MCP_GOOGLE_API_KEY

(none)

No

Google API key. Enables the Gemini provider (gemini-3.1-flash-image and others) when set. Get a key at https://aistudio.google.com/apikey.

IMAGE_GENERATION_MCP_SD_WEBUI_HOST

(none)

No

SD WebUI base URL (such as http://localhost:7860). Enables the SD WebUI provider when set. Compatible with AUTOMATIC1111, Forge, reForge, and Forge-neo.

IMAGE_GENERATION_MCP_SD_WEBUI_MODEL

(none)

No

SD WebUI checkpoint name, used for model-aware preset detection (SD 1.5 / SDXL / Lightning) and checkpoint override. Unset uses the instance's current model.

IMAGE_GENERATION_MCP_DEFAULT_PROVIDER

auto

No

Provider used when no keyword triggers auto-selection: auto, openai, gemini, sd_webui, or placeholder. auto picks the first configured provider.

IMAGE_GENERATION_MCP_TRANSFORM_CACHE_SIZE

64

No

Maximum number of transformed image results (resize, crop, convert) kept in memory. Set 0 to disable caching.

IMAGE_GENERATION_MCP_PAID_PROVIDERS

openai

No

Comma-separated provider names that cost money; generate_image asks for confirmation (client elicitation) before using them. An empty value falls back to this default; to disable confirmation, set a value that names no provider (such as none).

IMAGE_GENERATION_MCP_STYLES_DIR

~/.image-generation-mcp/styles

No

Directory for style preset files (Markdown with YAML front matter). Created automatically if it does not exist.

IMAGE_GENERATION_MCP_ALLOW_LOCAL_FILE_INPUT

false

No

Allow reading input images from local filesystem paths. Off by default: only URLs and uploads are accepted.

IMAGE_GENERATION_MCP_MAX_INPUT_IMAGE_BYTES

20971520

No

Maximum accepted input image size in bytes.

IMAGE_GENERATION_MCP_FETCH_TIMEOUT_S

30.0

No

HTTP timeout in seconds when fetching remote image URLs (fetch_image and URL inputs).

IMAGE_GENERATION_MCP_TRANSFER_TTL_DEFAULT_S

3600.0

No

Link lifetime in seconds when the caller requests no explicit TTL.

IMAGE_GENERATION_MCP_TRANSFER_TTL_MAX_S

86400.0

No

Ceiling in seconds a caller-requested link TTL is clamped to.

IMAGE_GENERATION_MCP_TRANSFER_GRACE_TTL_S

60.0

No

Post-success grace window in seconds: a served token's TTL shrinks to this so a stalled transfer can retry within it.

IMAGE_GENERATION_MCP_TRANSFER_LEASE_S

60.0

No

Crashed-handler reclaim window in seconds for an in-flight reservation.

IMAGE_GENERATION_MCP_TRANSFER_MAX_UPLOAD_BYTES

104857600

No

Maximum size in bytes of a single upload.

The create_download_link / create_upload_link tools and the /transfer/{token} route register only on an HTTP or SSE transport with BASE_URL set, and store link tokens in IMAGE_GENERATION_MCP_KV_STORE_URL; the IMAGE_GENERATION_MCP_TRANSFER_* knobs above tune link lifetime and upload limits. Security: IMAGE_GENERATION_MCP_ALLOW_LOCAL_FILE_INPUT grants callers server-filesystem read access via reference-image paths; enable it only for trusted callers or local single-user deployments.

Domain-config fields are composed inside src/image_generation_mcp/config.py between the CONFIG-FIELDS-START / CONFIG-FIELDS-END sentinels; env reads go through fastmcp_pvl_core.env(_ENV_PREFIX, "SUFFIX", default) so naming stays consistent, and field invariants go in __post_init__ between the CONFIG-VALIDATE-START / CONFIG-VALIDATE-END sentinels. Each field's metadata help and tags generate the table above directly, so keep them accurate and complete.

Key design decisions

  • Multi-provider with capability discovery, not feature flags. Each provider's discover_capabilities() reports its actual supported aspect ratios / qualities / formats / negative-prompt support at startup; routing logic asks the capability surface, not a hard-coded enum. New providers slot in by satisfying the protocol, with no router edits needed. (See docs/decisions/0001-…, 0002-…, 0007-….)

  • Per-model style_profile metadata, surfaced via list_providers. Closed-list providers (OpenAI, Gemini, placeholder) use exact-key lookup; SD WebUI uses a regex-ordered pattern table. Profiles include lifecycle flags (current / legacy / deprecated) and feed an auto-built top-level warnings array. (See docs/decisions/0009-….)

  • Hybrid background tasks. Short calls (OpenAI ~5 s) stream progress in-line; long calls (SD WebUI 30-180 s) run as background tasks with check_generation_status polling; clients pick the mode via task=True. (See docs/decisions/0005-….)

  • Image asset model: content-addressed registry + sidecar JSON metadata + on-demand transforms. Generated images keep their full-resolution original; image://{id}/view?format=webp&width=512&crop_x=… resources do format conversion / resize / crop on demand without re-generating. Transforms are cached. (See docs/decisions/0006-….)

  • Style library. User-saved markdown briefs (with YAML frontmatter for tags / aspect ratio / quality) that the LLM interprets per-provider, not copy-pasted verbatim. Distinct from per-model style_profile: style library is the brief; style_profile describes the model. (See docs/decisions/0008-… and 0009-… for disambiguation.)

  • Composes fastmcp_pvl_core.ServerConfig, never inherits. Domain config goes between CONFIG-FIELDS-START / CONFIG-FIELDS-END sentinels; env reads route through fastmcp_pvl_core.env(...) to keep prefix naming consistent.

Available Tools

10 tools
check_generation_statusCheck Generation StatusA
Read-onlyIdempotent

Check whether a background image generation has finished.

Call this after generate_image to wait for completion. Returns a short JSON status — no image data, no heavy UI.

  • "completed" → call show_image(uri=original_uri) to display the finished image.

  • "generating" → wait and check again.

  • "failed" → report the error to the user.

Inputs: image_id (required). Returns: result

edit_imageEdit ImageA
Read-only

Open an image for interactive editing (crop, rotate, flip).

The user edits in the viewer UI and saves as a new image. Always edits the original image — resource template transforms are ephemeral and LLM-facing; editor transforms are persistent and user-facing.

Inputs: image_id (required).

get_server_infoServer InfoA
Read-only

Report wrapper and upstream version info for image-generation-mcp. Returns server_name, server_version, core_version (fastmcp-pvl-core), and (when configured) an upstream version block. Useful for verifying a deployment matches the expected build.

Inputs: no named parameters. Returns: object

list_providersList ProvidersA
Read-only

List available image generation providers, models, and capabilities.

Returns provider names, available models, and capability details. Each model includes a prompt_style field: use "clip" for comma-separated CLIP tags (SD 1.5/SDXL) or "natural_language" for descriptive sentences (Flux, OpenAI). Each call includes a refreshed_at timestamp. Pass force_refresh=true if providers may have changed since the last check.

Also available as the info://providers resource for clients that support MCP resources.

Inputs: force_refresh. Returns: result

list_resourcesList ResourcesA
Read-only

List all available resources and resource templates.

Returns JSON with resource metadata. Static resources have a 'uri' field, while templates have a 'uri_template' field with placeholders like {name}.

Inputs: no named parameters. Returns: result

read_resourceRead ResourceA
Read-only

Read a resource by its URI.

For static resources, provide the exact URI. For templated resources, provide the URI with template parameters filled in.

Returns the resource content as a string. Binary content is base64-encoded.

Inputs: uri (required). Returns: result

show_imageShow ImageA
Read-only

Display a completed image with optional on-demand transforms.

Only call this for completed images — use check_generation_status to poll, then call this once when status is "completed".

Accepts a full image:// resource URI (e.g. image://abc123/view or image://abc123/view?format=webp&width=512). Transforms are encoded in the URI query string — no separate parameters needed.

Read the image://list resource to browse available image IDs.

Inputs: uri (required).

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.9/5.0

    Scored across 10 tools

    Disambiguation4/5

    Most tools have clearly distinct purposes (polling, displaying, browsing, editing, listing providers). However, gallery_page and gallery_full_image are app-only helpers that overlap conceptually with browse_gallery and show_image, creating mild ambiguity for an agent.

    Naming Consistency3/5

    The majority follow a verb_noun pattern (check_generation_status, show_image, browse_gallery, edit_image, list_providers, get_server_info, list_resources, read_resource), but gallery_page and gallery_full_image break the pattern by starting with a noun. This makes the naming convention somewhat inconsistent.

    Tool Count5/5

    With 10 tools, the count falls within the ideal 3–15 range and feels well-scoped for an image generation server, assuming the missing generate_image tool existed. The number itself is appropriate.

    Completeness1/5

    The server is named image-generation-mcp, yet there is no generate_image tool among the listed functions; check_generation_status references it but it is absent. This is a severe gap—the core generation capability is missing, leaving only supporting utilities and viewing/browsing tools.

    Maintenance

    ActivityStale
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      Not graded
      quality
      D
      maintenance
      MCP server for generating and editing images using OpenRouter API, with support for multiple models and dynamic model discovery. Runs in Docker and integrates with Claude Desktop via stdio transport.
      MIT