> ## Documentation Index
> Fetch the complete documentation index at: https://docs.introw.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Weekly Channel Slack Digest

> Use when a Channel Chief, Channel Ops, or RevOps user wants to generate the weekly partner-team Slack digest, wins, registrations, at-risk partners, pending approvals, KPI deltas,…

*Vendor skill for [Use case 14: Ecosystem Performance](/headless/agentic-use-cases/ecosystem-performance).*

**Skill ID:** `vendor-slack-weekly-channel-digest`

Use when a Channel Chief, Channel Ops, or RevOps user wants to generate the weekly partner-team Slack digest, wins, registrations, at-risk partners, pending approvals, KPI deltas, action items, and post it directly to the channel-team Slack channel. Trigger phrases include "weekly channel digest", "Monday partner update", "post the partner digest", "weekly Slack update", "channel team digest", "what happened with partners this week".

**Built for:** Partner Program Manager · Channel Ops Manager · Channel RevOps

**Workflow:** Trailing 7 days across systems → Wins · regs · at-risk · queue · KPI deltas · anomalies → First run preview · then auto-post → Channel-team Slack · message link logged

```md SKILL.md theme={"theme":{"light":"github-light","dark":"github-dark"}}
---
name: vendor-slack-weekly-channel-digest
description: Use when a Channel Chief, Channel Ops, or RevOps user wants to generate the weekly partner-team Slack digest, wins, registrations, at-risk partners, pending approvals, KPI deltas, action items, and post it directly to the channel-team Slack channel. Trigger phrases include "weekly channel digest", "Monday partner update", "post the partner digest", "weekly Slack update", "channel team digest", "what happened with partners this week".
---

# Weekly Channel Slack Digest (Vendor)

**Audience**: Vendor, uses **claude.ai Introw** MCP + **claude.ai Slack** MCP.
**Use case**: 14, Ecosystem Performance.

## When to use this skill

Use when a Channel Chief, Channel Ops, or RevOps user wants to generate the weekly partner-team Slack digest, wins, registrations, at-risk partners, pending approvals, KPI deltas, action items, and post it directly to the channel-team Slack channel.

**Sample prompts that fire this skill:**
- "weekly channel digest"
- "Monday partner update"
- "post the partner digest"
- "weekly Slack update"
- "channel team digest"
- "what happened with partners this week"

## Why this matters
Most channel teams write their Monday digest by hand: someone opens 5 tabs (CRM, PRM, finance, Slack threads, Linear), copies numbers, paragraphs them up, posts to `#channel-team`. It takes 60–90 minutes and gets skipped half the time. The information ages by Tuesday.

Agentic generation collapses prep to seconds. The digest stops being a chore (and stops getting skipped). The leadership team gets a consistent, current weekly heartbeat. Pair it with `vendor-anomaly-detector` to make the digest predictive, not just retrospective.

## What the digest contains (default sections)

1. **Wins**: closed-won partner-attached deals this week, with partner attribution.
2. **New registrations**: count, total ARR value, top deals.
3. **Movers**: partners pacing up or down notably vs. prior week (engagement, registration cadence, deal velocity).
4. **At risk**: partners who slipped this week (engagement decay, missed milestones, dormant). Top 3 with suggested intervention.
5. **In the queue**: pending approvals, MDF requests, deal-reg conflicts awaiting resolution. Aging items flagged.
6. **KPI snapshot**: activation rate, partner-attached pipeline, tier movement, week-over-week deltas.
7. **Anomalies**: anything unusual surfaced by `vendor-anomaly-detector` worth a human eye.
8. **Action items**: top 3–5 things the team should focus on this week, with owners.

User can configure which sections appear, the time window (default trailing 7 days), and which Slack channel to post to.

## Process

### Step 1: Resolve config
- Default time window: trailing 7 days, ending at run time.
- Default Slack channel: ask the user the first time, then capture.
- Section toggles: ask for any sections to skip (e.g., user only wants Wins + At risk).

### Step 2: Pull data
- `Introw:search_crm_objects`: closed-won deals + new opps in window (partner-attached).
- `Introw:search_form_submissions`: new registrations in window.
- `Introw:search_partner_engagement`: engagement deltas vs. prior period.
- `Introw:search_partners`: partner roster + lifecycle states.
- `Introw:search_tasks`: overdue / aging tasks across the team.
- `Introw:search_commissions`: commission events for context.
- `Introw:get_goals`: pacing vs. quarterly goals.
- Optionally: re-use output from `vendor-anomaly-detector` if it ran.

### Step 3: Compose the digest
Build a Slack-friendly message using Slack's block kit (or simple markdown if user prefers):
- Header block with brand + week-of date range.
- Section blocks for each enabled section.
- Use bullet lists with bold labels.
- Include partner names with hyperlinks back to the partner record in Introw where possible.
- Keep total length under ~3,000 characters, Slack truncates long messages.
- For long sections, summarize in the post + link to a full report.

### Step 4: Post (or surface for approval)
- **First run**: surface the drafted digest to the user for review. Confirm channel + tone. Don't auto-post.
- **After approval**: post via `Slack:*` to the configured channel.
- For recurring runs (e.g., weekly cron), the user can enable auto-post once the format is dialed in.
- Capture the posted message URL via `Introw:add_comment` on a designated "weekly digest log" record (or a custom CRM object) so the audit trail exists.

## Output format
- **The digest itself**: Slack-formatted message (block kit JSON or markdown).
- **Confirmation**: posted-to channel, message link, timestamp.
- **Variants** for different audiences if requested (e.g., a shorter exec version for CRO Slack).

## Guardrails & PRM best practice
- **Don't auto-post on first run.** Always preview + approve. Tone calibration matters, a digest that misreads sentiment ("celebrating" a partner who just churned) burns trust faster than no digest.
- **Be honest about losses.** If the week was bad, the digest should say so. Sugarcoated digests train the team to ignore them.
- **Names matter.** Cite partners and team members by name (with proper Slack mentions where relevant): generic "the team" digests get skimmed.
- **Action items must have owners.** "Need to follow up on X" with no name attached doesn't drive action; either assign or remove.
- **Aging is the alarm bell.** Items aging past SLA (approvals > 24h, conflicts unresolved > 48h) get flagged in red, this is the single most useful section for ops.
- **Respect channel volume.** If the channel is high-traffic, default to a shorter format and link out for detail. If it's quiet, the digest is the headline.
- **No partner-confidential data in public channels.** Commission specifics, strategic-account discussions, tier moves, none of those go in `#channel-team` if it's broadly accessible. Use a private channel for those if needed.
- **Don't post during incidents.** If the team is mid-fire (outage, major customer escalation), defer the digest. The skill should ask before posting if the user wants to skip this week.
- **Cross-skill handoff.** Anomaly section feeds from `vendor-anomaly-detector`; KPI snapshot is essentially the standing health check from `vendor-ecosystem-health-check` if installed; at-risk section can hand off to `vendor-activate-network-with-personalized-campaigns` for follow-up.
- **Iterate the format.** After 4 weeks, ask the team what they actually read. Cut the sections that don't get traction. The best digest is the one people open.
```

<Note>
  Drop this file into `.claude/skills/vendor-slack-weekly-channel-digest/SKILL.md` in your repo and Claude Code triggers it on the prompts in its `description`. Or run the same play in plain language from Claude, ChatGPT, Slack, Teams, or your CRM through [Introw's MCP server](/features/developer/mcp): every action writes back to your CRM source of truth.
</Note>
