drupal-mcp-connector 2.7.4 → 2.9.0

package/{.claude → .agents}/commands/drupal-get-taxonomy-terms.md

@@ -1,14 +1,13 @@
 ---
 description: "List all terms in a taxonomy vocabulary, sorted by name."
 argument-hint: "<vocabulary> [site] [limit] [offset]"
-allowed-tools: mcp__drupal__drupal_get_taxonomy_terms
 ---
 
-Call the `mcp__drupal__drupal_get_taxonomy_terms` MCP tool.
+Call the MCP tool `drupal_get_taxonomy_terms`.
 
 List all terms in a taxonomy vocabulary, sorted by name.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `vocabulary` (string): Vocabulary machine name, e.g. 'tags'
@@ -18,4 +17,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `limit` (number)
 - `offset` (number)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-get-user-by-name.md

@@ -1,14 +1,13 @@
 ---
 description: "Look up a Drupal user by their exact username."
 argument-hint: "<name> [site]"
-allowed-tools: mcp__drupal__drupal_get_user_by_name
 ---
 
-Call the `mcp__drupal__drupal_get_user_by_name` MCP tool.
+Call the MCP tool `drupal_get_user_by_name`.
 
 Look up a Drupal user by their exact username.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `name` (string): Drupal username (exact match)
@@ -16,4 +15,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-get-user.md

@@ -1,14 +1,13 @@
 ---
 description: "Fetch a single Drupal user account by UUID, including their assigned roles."
 argument-hint: "<id> [site]"
-allowed-tools: mcp__drupal__drupal_get_user
 ---
 
-Call the `mcp__drupal__drupal_get_user` MCP tool.
+Call the MCP tool `drupal_get_user`.
 
 Fetch a single Drupal user account by UUID, including their assigned roles.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `id` (string): User UUID
@@ -16,4 +15,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-governance-status.md

@@ -1,16 +1,15 @@
 ---
 description: "Report each configured site's source-governance condition. Always probes GET /drupal-mcp/readiness (even when this client does not require governance) and surfaces the server's reason verbatim. Never reports ok:true unless that check ran. Callable even while governed paths are denied — this is the diagnostic for that denial."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_governance_status
 ---
 
-Call the `mcp__drupal__drupal_governance_status` MCP tool.
+Call the MCP tool `drupal_governance_status`.
 
 Report each configured site's source-governance condition. Always probes GET /drupal-mcp/readiness (even when this client does not require governance) and surfaces the server's reason verbatim. Never reports ok:true unless that check ran. Callable even while governed paths are denied — this is the diagnostic for that denial.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-graphql-introspect.md

@@ -1,17 +1,16 @@
 ---
 description: "Introspect the Drupal GraphQL schema. Omit typeName for a full schema overview; provide typeName to get detailed fields and args for a specific type. Requires security.allowGraphql (off outside the development preset by default)."
 argument-hint: "[site] [typeName]"
-allowed-tools: mcp__drupal__drupal_graphql_introspect
 ---
 
-Call the `mcp__drupal__drupal_graphql_introspect` MCP tool.
+Call the MCP tool `drupal_graphql_introspect`.
 
 Introspect the Drupal GraphQL schema. Omit typeName for a full schema overview; provide typeName to get detailed fields and args for a specific type. Requires security.allowGraphql (off outside the development preset by default).
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `typeName` (string): Name of a specific GraphQL type to inspect in detail (optional)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-graphql.md

@@ -1,10 +1,9 @@
 ---
 description: "Execute a GraphQL query against a Drupal site. Requires the GraphQL Compose module (drupal.org/project/graphql_compose), which exposes a read-only schema. GraphQL is off unless security.allowGraphql is true (development preset only by default) because raw results bypass entity allowlists and field redaction. Mutations also require allowGraphqlMutations. Use drupal_graphql_introspect first to discover available types and fields. Example query: query GetArticle($id: String!) { nodeById(id: $id) { title ... on NodeArticle { body { value } } } } Example mutation (only if your GraphQL Compose schema enables mutations): mutation CreateArticle($title: String!, $body: String!) { createNodeArticle(data: { title: $title, body: { value: $body, format: \"full_html\" } }) { entity { title uuid } errors { message } } }"
 argument-hint: "<query> [site] [variables] [operationName]"
-allowed-tools: mcp__drupal__drupal_graphql
 ---
 
-Call the `mcp__drupal__drupal_graphql` MCP tool.
+Call the MCP tool `drupal_graphql`.
 
 Execute a GraphQL query against a Drupal site.
 Requires the GraphQL Compose module (drupal.org/project/graphql_compose), which
@@ -29,7 +28,7 @@ Example mutation (only if your GraphQL Compose schema enables mutations):
     }
   }
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `query` (string): GraphQL query or mutation string
@@ -39,4 +38,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `variables` (object (pass as JSON)): Variables to pass with the query
 - `operationName` (string): Operation name (for multi-operation documents)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-blocks.md

@@ -1,14 +1,13 @@
 ---
 description: "List custom content blocks (block_content entities), optionally scoped to a single block type (bundle). Returns each block's admin label (info) and body. Does not list code/plugin-defined blocks."
 argument-hint: "[site] [type] [limit] [offset] [sort]"
-allowed-tools: mcp__drupal__drupal_list_blocks
 ---
 
-Call the `mcp__drupal__drupal_list_blocks` MCP tool.
+Call the MCP tool `drupal_list_blocks`.
 
 List custom content blocks (block_content entities), optionally scoped to a single block type (bundle). Returns each block's admin label (info) and body. Does not list code/plugin-defined blocks.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
@@ -17,4 +16,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `offset` (number)
 - `sort` (array (pass as JSON)): Sort specs: [{ field, dir }]. Defaults to info asc.
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-content-types.md

@@ -1,16 +1,15 @@
 ---
 description: "List all content types defined on this Drupal site with their machine names and descriptions."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_list_content_types
 ---
 
-Call the `mcp__drupal__drupal_list_content_types` MCP tool.
+Call the MCP tool `drupal_list_content_types`.
 
 List all content types defined on this Drupal site with their machine names and descriptions.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-entity-types.md

@@ -1,16 +1,15 @@
 ---
 description: "Discover all JSON:API resource types (entity types + bundles) exposed by this Drupal site, filtered to only those your security config allows. Run this before working with an unfamiliar entity type."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_list_entity_types
 ---
 
-Call the `mcp__drupal__drupal_list_entity_types` MCP tool.
+Call the MCP tool `drupal_list_entity_types`.
 
 Discover all JSON:API resource types (entity types + bundles) exposed by this Drupal site, filtered to only those your security config allows. Run this before working with an unfamiliar entity type.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-media-types.md

@@ -1,16 +1,15 @@
 ---
 description: "List all media types defined on this Drupal site (image, document, remote_video, audio, etc.)."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_list_media_types
 ---
 
-Call the `mcp__drupal__drupal_list_media_types` MCP tool.
+Call the MCP tool `drupal_list_media_types`.
 
 List all media types defined on this Drupal site (image, document, remote_video, audio, etc.).
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-media.md

@@ -1,14 +1,13 @@
 ---
 description: "List media entities by type. Supports filtering by name substring and publish status."
 argument-hint: "[site] [type] [status] [name] [limit] [offset]"
-allowed-tools: mcp__drupal__drupal_list_media
 ---
 
-Call the `mcp__drupal__drupal_list_media` MCP tool.
+Call the MCP tool `drupal_list_media`.
 
 List media entities by type. Supports filtering by name substring and publish status.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
@@ -18,4 +17,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `limit` (number)
 - `offset` (number)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-menu-links.md

@@ -1,14 +1,13 @@
 ---
 description: "List custom (content) menu links, optionally scoped to a single menu (e.g. 'main', 'footer'). Returns each link's title, target URI, menu, and weight. Note: this lists editable menu_link_content entities, not code-defined static links."
 argument-hint: "[site] [menu] [limit] [offset] [sort]"
-allowed-tools: mcp__drupal__drupal_list_menu_links
 ---
 
-Call the `mcp__drupal__drupal_list_menu_links` MCP tool.
+Call the MCP tool `drupal_list_menu_links`.
 
 List custom (content) menu links, optionally scoped to a single menu (e.g. 'main', 'footer'). Returns each link's title, target URI, menu, and weight. Note: this lists editable menu_link_content entities, not code-defined static links.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
@@ -17,4 +16,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `offset` (number)
 - `sort` (array (pass as JSON)): Sort specs: [{ field, dir }]. Defaults to weight asc.
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-moderation-states.md

@@ -1,14 +1,13 @@
 ---
 description: "List the moderation states observed on a content type's content (best-effort; authoritative transitions require the Drush bridge)."
 argument-hint: "<type> [site] [sample]"
-allowed-tools: mcp__drupal__drupal_list_moderation_states
 ---
 
-Call the `mcp__drupal__drupal_list_moderation_states` MCP tool.
+Call the MCP tool `drupal_list_moderation_states`.
 
 List the moderation states observed on a content type's content (best-effort; authoritative transitions require the Drush bridge).
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `type` (string): Content type machine name
@@ -17,4 +16,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `sample` (number): How many recent items to sample
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-nodes.md

@@ -1,14 +1,13 @@
 ---
 description: "List nodes of a given content type. Supports status filtering, pagination, sorting, and structured filter descriptors."
 argument-hint: "<type> [site] [status] [limit] [offset] [filters] [sort]"
-allowed-tools: mcp__drupal__drupal_list_nodes
 ---
 
-Call the `mcp__drupal__drupal_list_nodes` MCP tool.
+Call the MCP tool `drupal_list_nodes`.
 
 List nodes of a given content type. Supports status filtering, pagination, sorting, and structured filter descriptors.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `type` (string): Content type machine name
@@ -21,4 +20,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `filters` (array (pass as JSON)): Structured filters: [{ field, op, value }]. op: eq|neq|gt|gte|lt|lte|contains|in|isNull
 - `sort` (array (pass as JSON)): Sort specs: [{ field, dir }] where dir is 'asc'|'desc'
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-revisions.md

@@ -1,14 +1,13 @@
 ---
 description: "Surface the addressable revisions of a content node: the latest default revision and the working-copy (forward) revision, with their version ids and links. workingCopy: null is not an all-clear — Drupal core can still reject PATCH when a revision row sits above the default without a content_moderation working copy (#201). The payload includes possiblyPatchBlocked (true when default changed is later than its revision_timestamp) plus changed and revisionTimestamp on latestVersion. Probe the host (this flag, then dryRun on the update) before creating dependent paragraphs. NOTE: JSON:API cannot enumerate full chronological revision history. Full history enumeration requires the Drush bridge."
 argument-hint: "<type> <id> [site]"
-allowed-tools: mcp__drupal__drupal_list_revisions
 ---
 
-Call the `mcp__drupal__drupal_list_revisions` MCP tool.
+Call the MCP tool `drupal_list_revisions`.
 
 Surface the addressable revisions of a content node: the latest default revision and the working-copy (forward) revision, with their version ids and links. workingCopy: null is not an all-clear — Drupal core can still reject PATCH when a revision row sits above the default without a content_moderation working copy (#201). The payload includes possiblyPatchBlocked (true when default changed is later than its revision_timestamp) plus changed and revisionTimestamp on latestVersion. Probe the host (this flag, then dryRun on the update) before creating dependent paragraphs. NOTE: JSON:API cannot enumerate full chronological revision history. Full history enumeration requires the Drush bridge.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `type` (string): Content type machine name, e.g. 'article'
@@ -17,4 +16,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-roles.md

@@ -1,16 +1,15 @@
 ---
 description: "List all user roles defined on this Drupal site."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_list_roles
 ---
 
-Call the `mcp__drupal__drupal_list_roles` MCP tool.
+Call the MCP tool `drupal_list_roles`.
 
 List all user roles defined on this Drupal site.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-sites.md

@@ -1,10 +1,9 @@
 ---
 description: "List the Drupal sites this principal may address. Each target includes the authoritative site name and base URL."
-allowed-tools: mcp__drupal__drupal_list_sites
 ---
 
-Call the `mcp__drupal__drupal_list_sites` MCP tool.
+Call the MCP tool `drupal_list_sites`.
 
 List the Drupal sites this principal may address. Each target includes the authoritative site name and base URL.
 
-This tool takes no arguments — call it directly (ignore `$ARGUMENTS`).
+This tool takes no arguments — call it directly.

package/{.claude → .agents}/commands/drupal-list-translations.md

@@ -1,14 +1,13 @@
 ---
 description: "List the translation langcode(s) for a Drupal entity (multilingual / content_translation). Reports the language(s) observable on the resource. Core JSON:API serves one language per resource and does not enumerate all translations — see the returned note. Defaults to node."
 argument-hint: "<type> <id> [site] [entityType]"
-allowed-tools: mcp__drupal__drupal_list_translations
 ---
 
-Call the `mcp__drupal__drupal_list_translations` MCP tool.
+Call the MCP tool `drupal_list_translations`.
 
 List the translation langcode(s) for a Drupal entity (multilingual / content_translation). Reports the language(s) observable on the resource. Core JSON:API serves one language per resource and does not enumerate all translations — see the returned note. Defaults to node.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Required:**
 - `type` (string): Bundle machine name, e.g. 'article'
@@ -18,4 +17,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `entityType` (string): Entity type machine name. Default: 'node'.
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-users.md

@@ -1,14 +1,13 @@
 ---
 description: "List Drupal user accounts. Filter by active/blocked status or by role machine name."
 argument-hint: "[site] [status] [role] [limit] [offset]"
-allowed-tools: mcp__drupal__drupal_list_users
 ---
 
-Call the `mcp__drupal__drupal_list_users` MCP tool.
+Call the MCP tool `drupal_list_users`.
 
 List Drupal user accounts. Filter by active/blocked status or by role machine name.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
@@ -17,4 +16,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `limit` (number)
 - `offset` (number)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-list-vocabularies.md

@@ -1,16 +1,15 @@
 ---
 description: "List all taxonomy vocabularies defined on this Drupal site."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_list_vocabularies
 ---
 
-Call the `mcp__drupal__drupal_list_vocabularies` MCP tool.
+Call the MCP tool `drupal_list_vocabularies`.
 
 List all taxonomy vocabularies defined on this Drupal site.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-mcp-whoami.md

@@ -1,16 +1,15 @@
 ---
 description: "Report the agent's effective governance tier, security preset, OAuth scopes, and capabilities (read/write/delete/config/publish) for a site. No credentials, no backend call."
 argument-hint: "[site]"
-allowed-tools: mcp__drupal__drupal_mcp_whoami
 ---
 
-Call the `mcp__drupal__drupal_mcp_whoami` MCP tool.
+Call the MCP tool `drupal_mcp_whoami`.
 
 Report the agent's effective governance tier, security preset, OAuth scopes, and capabilities (read/write/delete/config/publish) for a site. No credentials, no backend call.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-report-404-log.md

@@ -1,17 +1,16 @@
 ---
 description: "Aggregate Drupal 'page not found' (404) log events into the top missing URLs ranked by hit count — the redirect-candidate list. Self-sufficient via the connector's drush watchdog bridge; returns an 'unavailable' payload when drush isn't configured for the site."
 argument-hint: "[site] [limit]"
-allowed-tools: mcp__drupal__drupal_report_404_log
 ---
 
-Call the `mcp__drupal__drupal_report_404_log` MCP tool.
+Call the MCP tool `drupal_report_404_log`.
 
 Aggregate Drupal 'page not found' (404) log events into the top missing URLs ranked by hit count — the redirect-candidate list. Self-sufficient via the connector's drush watchdog bridge; returns an 'unavailable' payload when drush isn't configured for the site.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `limit` (number): Max distinct missing paths to return (max 200)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-report-accessibility-audit.md

@@ -1,18 +1,17 @@
 ---
 description: "Accessibility audit for body content: images without alt text, inline H1 tags, non-descriptive link text ('click here', 'read more'), and tables without captions."
 argument-hint: "[site] [type] [sampleSize]"
-allowed-tools: mcp__drupal__drupal_report_accessibility_audit
 ---
 
-Call the `mcp__drupal__drupal_report_accessibility_audit` MCP tool.
+Call the MCP tool `drupal_report_accessibility_audit`.
 
 Accessibility audit for body content: images without alt text, inline H1 tags, non-descriptive link text ('click here', 'read more'), and tables without captions.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `type` (string): Content type (default: article)
 - `sampleSize` (number)
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-report-alias-coverage.md

@@ -1,18 +1,17 @@
 ---
 description: "URL-alias coverage for a content type: nodes whose canonical URL is still /node/N (no alias / pathauto gap), plus conflicting aliases (one alias mapped to multiple system paths) when the path_alias entity is exposed."
 argument-hint: "[site] [type] [sampleSize]"
-allowed-tools: mcp__drupal__drupal_report_alias_coverage
 ---
 
-Call the `mcp__drupal__drupal_report_alias_coverage` MCP tool.
+Call the MCP tool `drupal_report_alias_coverage`.
 
 URL-alias coverage for a content type: nodes whose canonical URL is still /node/N (no alias / pathauto gap), plus conflicting aliases (one alias mapped to multiple system paths) when the path_alias entity is exposed.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `type` (string): Content type (default: article)
 - `sampleSize` (number): Max nodes to scan
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-report-broken-embeds.md

@@ -1,18 +1,17 @@
 ---
 description: "Scan published body content for embedded entities (media/entity embeds) and report usage by type, flagging embeds with a missing/malformed data-entity-uuid. Full target-existence verification is gated on a future server-tool."
 argument-hint: "[site] [type] [sampleSize]"
-allowed-tools: mcp__drupal__drupal_report_broken_embeds
 ---
 
-Call the `mcp__drupal__drupal_report_broken_embeds` MCP tool.
+Call the MCP tool `drupal_report_broken_embeds`.
 
 Scan published body content for embedded entities (media/entity embeds) and report usage by type, flagging embeds with a missing/malformed data-entity-uuid. Full target-existence verification is gated on a future server-tool.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
 - `type` (string): Content type (default: article)
 - `sampleSize` (number): Max nodes to scan
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.

package/{.claude → .agents}/commands/drupal-report-broken-links.md

@@ -1,14 +1,13 @@
 ---
 description: "Inventory links in published body content (internal/external/images), aggregate external hosts, and flag malformed hrefs. With checkLive:true, verifies links via bounded, SSRF-guarded outbound HTTP (internal always; external only if includeExternal and host-allowlisted). No network egress unless checkLive is set."
 argument-hint: "[site] [type] [sampleSize] [checkLive] [includeExternal]"
-allowed-tools: mcp__drupal__drupal_report_broken_links
 ---
 
-Call the `mcp__drupal__drupal_report_broken_links` MCP tool.
+Call the MCP tool `drupal_report_broken_links`.
 
 Inventory links in published body content (internal/external/images), aggregate external hosts, and flag malformed hrefs. With checkLive:true, verifies links via bounded, SSRF-guarded outbound HTTP (internal always; external only if includeExternal and host-allowlisted). No network egress unless checkLive is set.
 
-Parse the request in `$ARGUMENTS` into this tool's parameters:
+Parse the arguments supplied with this command into this tool's parameters:
 
 **Optional:**
 - `site` (string): Named site from connector config. Omit only on reads: multi-site configs fall back to defaultSite (often local/dev, not production). Writes require an explicit site when more than one site is configured. Every response includes `_target` { name, baseUrl, source } (`hint` when you passed site, `default` when you did not).
@@ -17,4 +16,4 @@ Parse the request in `$ARGUMENTS` into this tool's parameters:
 - `checkLive` (boolean (true/false)): Perform live HTTP checks (off by default)
 - `includeExternal` (boolean (true/false)): When checkLive, also check allowlisted external hosts
 
-If a required parameter is missing from `$ARGUMENTS`, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.
+If a required parameter is missing, ask before calling — do not invent values. Coerce each value to its JSON type (booleans → true/false, numbers → numeric, object/array → parse JSON), then make the single tool call and summarize the result.