digest

Digest Command

If you see unfamiliar placeholders or need to check which tools are connected, see CONNECTORS.md.

Scan recent activity across all connected sources and generate a structured digest highlighting what matters.

Lark-native execution (depth core: LARK-PATTERNS, LARK-RECIPES, LARK-FUSION). Gather activity by fanning out the lark_* reads in parallel, each with a jq projection (P3); pull meeting action items from lark_minutes_search instead of re-deriving them (P6); resolve any name to an open_id with lark_contact_search (P1). Deliver the digest as an interactive card via lark_im_card_send (P4) — action items as item rows with side buttons, decisions/updates in collapsible panel sections — not a wall of text. A daily digest overlaps heavily with the installed daily-digest skill, and a start-of-day version with morning-brief; prefer delegating to those for the polished card, using this command's structure as the outline.

Instructions

1. Parse Flags

Determine the time window from the user's input:

• --daily — Last 24 hours (default if no flag specified)
• --weekly — Last 7 days

The user may also specify a custom range:

• --since yesterday
• --since Monday
• --since 2025-01-20

2. Check Available Sources

On Lark every source resolves to the lark MCP server (same mapping as the search command):

• ~~chat — lark_im_search (groups, DMs, mentions)
• ~~email — lark_api mail search (inbox, threads)
• ~~cloud storage — lark_doc_search (recently modified docs shared with user)
• ~~project tracker — lark_task_my, lark_base_search (tasks assigned/completed, tracker rows)
• ~~CRM — CRM Base via lark_base_search (opportunity/account activity, P5)
• ~~knowledge base — lark_doc_search (updated Wiki pages), lark_minutes_search (new meeting artifacts, P6)
• Also useful: lark_calendar_agenda (today's/upcoming events)

If the lark MCP server isn't authenticated, guide the user:

To generate a digest, the lark MCP server needs to be authenticated.
Run `auth login` (see the lark-shared skill) to connect your workspace.

3. Gather Activity from Each Source

Fire these in one batched turn, each with a jq projection scoped to the time window (P3):

~~chat — lark_im_search:

• Messages mentioning the user; threads the user participated in; new messages in key groups.
• jq=".data.messages[] | {chat, sender, text, create_time}".

~~email — lark_api mail search:

• Recent inbox messages; threads with new replies; mails with action items / questions for the user.

~~cloud storage — lark_doc_search:

• Docs recently modified or shared with the user; project with jq to title/url/edit_time.

~~project tracker — lark_task_my + lark_base_search:

• lark_task_my(jq=".data.items[] | {id, summary, due, completed}") (new/updated/due-soon).
• Tracker Base rows changed in the window via lark_base_search.

~~CRM — CRM Base (lark_base_search, P5):

• Opportunity stage changes; new activity on owned accounts; updated contacts — jq to the changed fields.

Meetings — lark_minutes_search (P6):

• lark_minutes_search(participant_ids="me", query=..., jq=".data.items[] | {title, summary, action_items}") — pull AI summaries/action items directly; don't re-summarize. lark_vc_search for past-meeting envelopes.

~~knowledge base — lark_doc_search:

• Recently updated Wiki docs / new docs in watched areas.

Calendar — lark_calendar_agenda:

• Today's/recent events for the time-window framing.

4. Identify Key Items

From all gathered activity, extract and categorize:

Action Items:

• Direct requests made to the user ("Can you...", "Please...", "@user")
• Tasks assigned or due soon
• Questions awaiting the user's response
• Review requests

Decisions:

• Conclusions reached in threads or emails
• Approvals or rejections
• Policy or direction changes

Mentions:

• Times the user was mentioned or referenced
• Discussions about the user's projects or areas

Updates:

• Status changes on projects the user follows
• Document updates in the user's domain
• Completed items the user was waiting on

5. Group by Topic

Organize the digest by topic, project, or theme rather than by source. Merge related activity across sources:

## Project Aurora
- IM: Design review thread concluded — team chose Option B (design group, Tuesday)
- Mail: Sarah sent updated spec incorporating feedback (Wednesday)
- Docs: "Aurora API Spec v3" updated by Sarah (Wednesday)
- Tasks: 3 tasks moved to In Progress, 2 completed
- Minutes: "Aurora Sync" AI summary — owner assigned for rollout (Tuesday)

## Budget Planning
- Mail: Finance team requesting Q2 projections by Friday
- IM: Todd shared template in finance group (Monday)
- Docs: "Q2 Budget Template" shared with you (Monday)

6. Format and Deliver the Digest

Deliver as an interactive card (P4) via lark_im_card_send — this is the Lark superpower for a digest. Structure: a colored header ("Daily Digest — "), an Action Items section as item rows each with a side button (e.g. Open / Complete → lark_task_complete), Decisions/Updates in collapsible panel sections, a note footer with the summary stats. Always print_json: true to validate the spec, then dry_run: true, then send (P2). Card grammar: the lark-im skill (references/spec-reference.md). For the polished implementation, delegate to daily-digest / morning-brief. Fall back to the markdown layout below only if the user wants it inline:

# [Daily/Weekly] Digest — [Date or Date Range]

Sources scanned: IM, Mail, Docs, Tasks, Minutes [others]

## Action Items (X items)
- [ ] [Action item 1] — from [person], [source] ([date])
- [ ] [Action item 2] — from [person], [source] ([date])

## Decisions Made
- [Decision 1] — [context] ([source], [date])
- [Decision 2] — [context] ([source], [date])

## [Topic/Project Group 1]
[Activity summary with source attribution]

## [Topic/Project Group 2]
[Activity summary with source attribution]

## Mentions
- [Mention context] — [source] ([date])

## Documents Updated
- [Doc name] — [who modified, what changed] ([date])

7. Handle Unavailable Sources

If any source fails or is unreachable:

Note: Could not reach [source name] for this digest.
The following sources were included: [list of successful sources].

Do not let one failed source prevent the digest from being generated. Produce the best digest possible from available sources.

8. Summary Stats

End with a quick summary:

---
[X] action items · [Y] decisions · [Z] mentions · [W] doc updates
Across [N] sources · Covering [time range]

Notes

• Default to --daily if no flag is specified
• Group by topic/project, not by source — users care about what happened, not where it happened
• Action items should always be listed first — they are the most actionable part of a digest
• Deduplicate cross-source activity (same decision in IM and mail = one entry)
• For weekly digests, prioritize significance over completeness — highlight what matters, skip noise
• If the user has a memory system (CLAUDE.md), use it to decode people names and project references
• Include enough context in each item that the user can decide whether to dig deeper without clicking through
• Deliver as an interactive card (lark_im_card_send, P4); pull meeting items from lark_minutes_search (P6); for a polished daily/start-of-day card, delegate to daily-digest / morning-brief