Skip to main content
Glama
Sign In
immich-mcp

immich-mcp

by si-essence
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Local

immich_people

DestructiveIdempotent

Manage Immich people and faces: create, merge, reassign, update, delete, and retrieve people, faces, statistics, thumbnails, and user access.

Instructions

immich_people covers the people toolset (19 operations). Auth header x-api-key. Writes need confirm:true and IMMICH_ALLOW_WRITES=true. Omit confirm for a dry-run preview. Actions: create_face: POST /faces. Create a face create_person: POST /people. Create a person delete_face: DELETE /faces/{id}. Delete a face delete_people: DELETE /people. Delete people delete_person: DELETE /people/{id}. Delete person get_all_people: GET /people. Get all people get_faces: GET /faces. Retrieve faces for asset get_person: GET /people/{id}. Get a person get_person_statistics: GET /people/{id}/statistics. Get person statistics get_person_thumbnail: GET /people/{id}/thumbnail. Get person thumbnail get_users_for_people: GET /people/users. Get people access merge_people: POST /people/merge. Merge people merge_person_legacy: POST /people/{id}/merge. Merge people reassign_faces: PUT /people/{id}/reassign. Reassign faces reassign_faces_by_id: PUT /faces/{id}. Re-assign a face to another person remove_users_from_people: DELETE /people/users. Remove users from people update_people: PUT /people. Update people update_person: PUT /people/{id}. Update person upsert_people_users: PUT /people/users. Upsert user access

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "action": "<action>"
    },
    "name": "immich_people"
  }
}

Input Schema

NameRequiredDescriptionDefault
bodyNoJSON body for the Immich operation. Secrets are redacted in output.
queryNoOpenAPI query parameters.
actionYesImmich operation to call.
paramsNoOpenAPI path parameters, such as id.
confirmNoSet to the boolean true to execute the write. Omit this field for a dry-run preview. false and the string true are rejected.
checksumNoOptional x-immich-checksum header for asset upload.
file_pathNoLocal file to upload as multipart assetData. Confirm-gated. The path is omitted from output.
save_pathNoLocal path to save a binary download. If omitted, only metadata is returned. The path is omitted from output.
sidecar_pathNoOptional local XMP sidecar for asset upload. The path is omitted from output.

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

A3.5/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare destructiveHint=true, openWorldHint=true, idempotentHint=true and readOnlyHint=false; the description earns credit for adding the x-api-key auth header, the two-part write gate (confirm + env flag), and the dry-run-when-omitted contract, all of which are non-obvious from the annotations. However, it does not identify which of the 19 actions are destructive, whether deletes cascade to face data, or whether merges are reversible — the most important behavioral facts for a destructive multiplexer are left out.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The first three sentences are front-loaded and high-value: scope, auth, and the write/confirm gate. The remainder is a 19-entry list where each entry repeats the enum name verbatim ('create_person: POST /people. Create a person') — near-zero information beyond the schema enum, which bloats the description substantially.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a 19-operation multiplexer with 9 parameters and no output schema, the definition covers scope, auth and the write gate adequately. It leaves real gaps: no per-action parameter requirements, no indication of what each action returns, and no guidance on the overlapping pairs (reassign_faces vs reassign_faces_by_id, merge_people vs merge_person_legacy), which an agent cannot resolve from the description alone.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the schema already explains body, query, params, confirm, checksum, file_path, save_path and sidecar_path; baseline is 3. The per-action glosses map each action to an endpoint (DELETE /faces/{id}, POST /people/merge), which adds a little routing meaning, but they are essentially restatements of the enum values and never indicate which action needs which param (e.g. that get_person needs params.id).

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb-resource pairing: it 'covers the people toolset (19 operations)' and enumerates every action with its HTTP verb and endpoint, so an agent knows this is the people/face management surface. It never names the sibling tools (immich_albums, immich_assets, immich_search, immich_server) to draw the boundary explicitly, so differentiation is by domain noun only.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives concrete operational guidance for how to invoke: writes require confirm:true plus IMMICH_ALLOW_WRITES=true, and omitting confirm yields a dry-run preview. That is a clear policy for the risky path. It stops short of saying when to prefer this tool over a sibling (e.g. people search vs immich_search) or which action to pick for a given user intent.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server

Other Tools