Attach an AI-generated review comment to an eLabFTW experiment or item. The provenance header ('[AI-GENERATED] | | trace: …') is added automatically. WRITE OPERATION — requires the comment write scope.

Add a new step (task) to an existing eLabFTW experiment, item, experiment template or item type. WRITE OPERATION — requires the step write scope. A deadline (date/datetime string) is set with a follow-up update because the create endpoint only stores the body. Returns entity_type, entity_id, step_id, the stored deadline and status='created'.

Persist a list of (suggested) tags to an eLabFTW experiment or item. WRITE OPERATION — requires the tag write scope.

Idempotently remove several directed links in one call. 'pairs_json' is a JSON array of {from_type, from_id, to_type, to_id} objects. WRITE OPERATION when dry_run=false — requires the link write scope. Returns a per-pair result list plus a summary.

Idempotently create several directed links in one call. 'pairs_json' is a JSON array of {from_type, from_id, to_type, to_id} objects, at most bulk_links_max_pairs entries. WRITE OPERATION when dry_run=false — requires the link write scope. Returns a per-pair result list plus a summary.

Create a new eLabFTW experiment. WRITE OPERATION — requires the create write scope. Discover valid category/status IDs first; never embed IDs in the title or body. If the team enforces experiment templates (force_exp_tpl), this tool is refused and create_experiment_from_template must be used. Returns the new id.

Create a new eLabFTW experiment from an experiment template. WRITE OPERATION — requires the create write scope. Use list_experiment_templates to find template IDs. include_steps/include_body decide whether the template's steps and body are copied. Returns the new id.

Create a new eLabFTW database item (resource). WRITE OPERATION — requires the create write scope. Items belong to item categories; discover valid IDs first and never embed IDs in the title or body. Returns the new id.

Idempotently remove a directed link between an experiment and an item. WRITE OPERATION when dry_run=false — requires the link write scope. dry_run=true (default) only reports. Returns result: unlinked | already_absent | would_unlink.

Delete a step from an eLabFTW experiment, item, experiment template or item type. WRITE OPERATION — requires the step write scope. Returns status='deleted'.

Idempotently create a directed link between an experiment and an item. WRITE OPERATION when dry_run=false — requires the link write scope. dry_run=true (default) only reports what would happen. Returns result: linked | already_linked | would_link (verified after writing).

Resolve the target entity by query and then link it idempotently. Query resolution works for experiments and items only. WRITE OPERATION when dry_run=false — requires the link write scope. With several matches the tool returns the candidates instead of guessing (on_ambiguous='error' makes it fail instead).

Build the directed link graph around a root experiment/item via BFS. direction is 'both', 'outgoing' or 'incoming'; max_depth is capped at 3; node/edge/fan-out caps and optional filters (allowed_types, category_ids, status_ids, title_regex as JSON arrays/pattern) keep the graph bounded. Returns root, parameters, nodes, edges and truncation flags. Read-only.

Return connection and session details for the active server: eLabFTW base URL, authenticated user, active team, effective server mode (read-only vs write-enabled), enabled tool groups, and a category catalogue (experiments and items) for the active team. Always safe to call; never writes anything.

Return a sanitized subset of the current user's capabilities and team roles. Calls /users/me and returns only the fields relevant for capability detection: userid, fullname, current team id, team memberships (id, name, is_admin, is_owner, is_archived), is_sysadmin, can_manage_compounds and can_manage_inventory_locations. Sensitive fields (email, API keys, signatures, preferences) are deliberately omitted.

Fetch directed link edges of a single eLabFTW experiment or item: outgoing links (items_links / experiments_links) and incoming ones (related_items_links / related_experiments_links). Each edge carries from {type,id}, to {type,id} and a direction label. Read-only.

Fetch a single elabFTW experiment by its numeric ID. Returns the full experiment record including body, metadata, tags, comments, links and uploads. Always safe to call; never writes anything.

Fetch a single elabFTW experiment template by its numeric ID. Returns the full template record including body, metadata and steps. Use list_experiment_templates to discover templates and create_experiment_from_template to create an experiment from one. Always safe to call; never writes anything.

Fetch a single elabFTW database item (resource) by its numeric ID. Returns the item with metadata, tags, comments, links and uploads. Always safe to call; never writes anything.

Fetch a single elabFTW item type by its numeric ID. Returns the full item type record including body, metadata and steps. Use list_item_types to discover item types. Always safe to call; never writes anything.

List all experiment categories for the active elabFTW team. Specify team_id='current' (default) or a numeric team ID. Returns id, title, color and is_default per category — use those ids when filtering list_experiments by category. Always safe to call; never writes anything.

Search or list elabFTW experiments. Returns a structured response with paging metadata, applied filters, and result records. Check paging.returned_count vs paging.limit to detect further pages. Use the dedicated filter arguments for known category/user/status IDs or tags instead of putting them into q. Always safe to call; never writes anything.

List all experiment statuses for the active eLabFTW team. Specify team_id='current' (default) or a numeric team ID. Read-only.

List elabFTW experiment templates (/experiments_templates). Returns a structured response with paging metadata and result records. Templates are reusable pre-filled starting points for new experiments and are DIFFERENT from item types (use list_item_types for those). Use get_experiment_template for a single template and create_experiment_from_template to use one. Always safe to call; never writes anything.

List all item categories (resource categories) for the active eLabFTW team. Specify team_id='current' (default) or a numeric team ID. Use the returned ids when filtering list_items by category. Always safe to call; never writes anything.

Search or list elabFTW database items (resources). Returns a structured response with paging metadata, applied filters, and result records. Check paging.returned_count vs paging.limit to detect further pages. Use the dedicated filter arguments for known IDs or tags instead of q. Always safe to call; never writes anything.

List all item statuses (resource statuses) for the active eLabFTW team. Specify team_id='current' (default) or a numeric team ID. Read-only.

List elabFTW item types (/items_types), also called resource categories. They define the category structure and default content for database items and are DIFFERENT from experiment templates (use list_experiment_templates). Use get_item_type for a single one. Always safe to call; never writes anything.

List all steps (tasks) attached to an eLabFTW experiment, item, experiment template or item type. Returns entity_type, entity_id, total_steps and a steps array (id, body, finished, ordering, deadline). With render_markdown=True a 'markdown' field with a GFM checkbox task list and deep links to each step is added. Read-only.

Refresh and return the cached team capability flags for the current session. Fetches the team record, sanitizes the capability fields used for policy checks, updates the cache and returns the sanitized team_caps object. Does not write to eLabFTW resources.

Search experiments or items by query and return compact candidates for disambiguation. Read-only, always available. Returns status, entity_type, resolution (not_found | unique | ambiguous) and the candidate list — use it to find an id before acting on it.

Generate an AI review summary for an eLabFTW experiment (documentation quality, risks, suggestions, completeness score). Read-only — writes nothing. Needs an AI endpoint; without one the response carries the upstream placeholder marker.

Suggest structured metadata (custom fields) for an eLabFTW experiment or item. Read-only — writes nothing; feed the result into update_entity_metadata to persist it.

Suggest tags for an eLabFTW experiment or item based on its content. Read-only — writes nothing; use apply_tag_suggestions to persist them.

Mark a step as complete or incomplete. WRITE OPERATION — requires the step write scope. finished=1 marks it complete, finished=0 incomplete; when omitted the server-side finish action flips the current state.

Update top-level scalar fields (title, status, category) on an existing eLabFTW experiment or item. WRITE OPERATION — requires the update write scope. At least one of title, status_id, category_id must be given. Returns the applied fields and a read-back verification.

Update the metadata (custom fields) of an eLabFTW experiment, item, experiment template or item type. WRITE OPERATION — requires the update write scope. 'fields' is a JSON string with an array of field objects ({key, value | values, type, group_id, allow_multi_values, description}); 'groups' optionally defines extra field groups ([{id, name}]). mode='merge' (default) keeps the existing metadata, 'overwrite' rebuilds it from the supplied fields. Returns the resulting field names and a read-back verification.

Update the body of an existing eLabFTW experiment. WRITE OPERATION — requires the update write scope. The existing content_type (HTML=1 / Markdown=2) is preserved. mode='append' (default) keeps the original body byte-for-byte and appends the new text; mode='overwrite' replaces it and needs both the config flag allow_body_overwrite and confirm_overwrite=true. The result reports a read-back verification of the change.

Update the body of an existing eLabFTW database item. WRITE OPERATION — requires the update write scope. mode='append' (default) preserves the original body and appends; mode='overwrite' replaces it and needs allow_body_overwrite plus confirm_overwrite=true.

Update an existing step: body and/or deadline. WRITE OPERATION — requires the step write scope. At least one of body, deadline, deadline_notif must be provided; deadline_notif is reported as unsupported when the API rejects it (see notes).

Upload a single image attachment to an eLabFTW experiment or item from base64-encoded content. WRITE OPERATION — requires the update write scope. Only image/png, image/jpeg, image/gif and image/tiff are accepted. Returns the new upload id and size.