Operate Feishu Bitable tables, fields, records, views, dashboards, workflows, and roles via lark-cli.
npx skills add https://github.com/U1traVeno/skills --skill lark-base-u1traveno
Please help me install this Agent Skill. Skill: lark-base Source: https://github.com/U1traVeno/skills/tree/main/profiles/lark/lark-base Command: npx skills add https://github.com/U1traVeno/skills --skill lark-base-u1traveno
lark-cli base +... shortcut with the right references, so tables, fields, records, views, forms, dashboards, workflows, and role permissions are handled correctly the first time. ## Core Features & Use Cases - Table, Field & Record Operations: Create and copy bases, manage tables and fields (including formula and lookup fields), and read or write records with correct CellValue formats, batch limits, and attachment commands. - Query & Aggregation: Run server-side filtering, grouping, aggregation, and TopN analysis with +data-query, plus view filters and record search for row-level results. - Forms, Dashboards, Workflows & Roles: Submit shared forms, build dashboard chart blocks with validated data_config, manage automation workflows, and configure advanced permissions and custom roles. - Use Case: A user shares a /base/ link and asks for a sales dashboard. The Skill resolves the base token, reads the real table schema, creates a dashboard, and adds chart blocks one by one using the data_config SSOT. ## Quick Start Ask the assistant to use the lark-base skill to list the tables and fields in the Feishu Bitable behind your /base/ link, then create or update records as needed.High-intent search queries and answers about installing and using this skill.
Use lark-cli base +record-upsert, +record-batch-create, or +record-batch-update after reading the field list. Cell values must match each field type, batches are capped at 200 records, and formula, lookup, and system fields are read-only.
Use lark-cli base +data-query with a JSON DSL defining dimensions, measures, filters, sort, and pagination. Aggregation runs server-side in the Base query service, so global counts and TopN results are not limited by local record-list pagination.
Read the formula-field-guide or lookup-field-guide reference first, then call +field-create or +field-update with a JSON body containing the expression and the --i-have-read-guide flag. Table and field names in the expression must exactly match the real schema.
No. File imports are handled by lark-drive using lark-cli drive +import --type bitable. After the import completes, you return to lark-base commands to manage the resulting tables and records.
The invalid token error usually means a wiki token, workspace token, or full URL was passed as --base-token; resolve it with +url-resolve or +title-resolve first. Error 91403 means no access to the Base and should be handled through the lark-shared permission flow, not retried.
Extract the share token from the /share/base/form/ link, run +form-detail to read question types, required flags, and filters, then call +form-submit with a fields JSON. Attachments go in a separate attachments object and require the form's base_token.
augustsclOperate Feishu Base tables, fields, records, views, and workflows via lark-cli.
ojowwalker77Display Matrix memory status and export recent solutions as JSON.
baibairuiAutomate Feishu/Lark cloud drive file and document operations via lark-cli.
SpaceZephyrParse Feishu/Lark OpenAPI markdown docs to locate and invoke native endpoints via lark-cli.