Find why your productivity system keeps failing, then apply the smallest fix — capacity math, bottleneck routing, durable local notes.
Documents
Lark Work Report
Try itPreflight-check Lark channel and send completion reports
What it does
Use this skill when the user asks to: report completion/progress to Feishu/Lark establish or repair a stable "work group" reporting channel before task execution
The skill document
Lark Work Report
Use this skill when the user asks to:
- report completion/progress to Feishu/Lark
- establish or repair a stable "work group" reporting channel before task execution
Required MCP interfaces
- Find group:
im_v1_chat_list(useUAT=true) - Send report:
im_v1_message_create(useUAT=true,receive_id_type=chat_id) - Create work group if needed:
im_v1_chat_create(useUAT=true) - Resolve user identity when needed:
contact_v3_user_batchGetId(useUAT=true)
Default work-group bootstrap
Target group name: Codex工作群
- Run
im_v1_chat_listand search exact group name match (name == "Codex工作群"). - If exactly one match exists, store its
chat_idand send a short connectivity test message. - If no exact match:
- list likely candidates from
im_v1_chat_list(names containingCodexor工作群) - ask user to confirm whether one candidate should be used
- list likely candidates from
- If still unresolved:
- ask user for identity details (name/mobile/email/open_id)
- ask if a new
Codex工作群should be created - create group via
im_v1_chat_create, send a test message, and wait for user confirmation
- Send one test message to the selected/new group and confirm with the CLI user that the channel is correct.
- Only after reporting channel is confirmed, proceed with the main task workflow.
Reporting workflow
After finishing a user task:
- Send a concise report message to the resolved work-group
chat_id. - Report should include:
- task summary
- key outputs/artifacts
- blockers/risks (if any)
- next-step suggestion (if any)
- Keep message text compact. If content is long, split into multiple sequential messages.
Explicit non-goals
- This skill does not monitor Feishu for follow-up instructions.
- This skill does not run polling loops for new tasks.
- Task intake/clarification happens in CLI conversation unless the user explicitly requests another workflow.
Safety and behavior constraints
- Never assume group identity from memory; always verify via API in the current session.
- If sender identity or group identity is ambiguous, confirm with the user before sending task reports.
- Do not silently switch to another group without explicit user confirmation.
Related skills
Generate and edit Draw.io, Mermaid, and Excalidraw diagrams from natural language using a structured JSON spec.
Join a video meeting as an AI bot with voice, avatar, and screenshare across four operating modes.
Stores durable facts in a categorized, plain-markdown vault on disk, alongside your agent's built-in memory.
Fetch raw ad creative, app, ranking, and revenue data from AdMapix as structured JSON.
Write, debug, and tune Playwright specs with locator strategy, trace diagnosis, and CI-aware timeouts.
More from xukp20
Browse all skillsModel-led arXiv collection: author queries, judge relevance per result, then merge and dedupe into a final paper set.
Download ArXiv source or PDF artifacts per paper, read the paper, and write a structured summary.md in your chosen language.
Orchestrates end-to-end arXiv paper retrieval, processing, and batch reporting with language control and parallel or serial paper handling modes.
Generates a batch arXiv report by combining model-written templates with brief conclusions and abs URLs extracted from per-paper summary.md files.
Review designs and plans through a focused convergence loop
Organize auditable task packages and launch workflows