octavehq/signals

/octave:signals - Morning Intelligence Briefing

Your daily command center. Surfaces the deals that moved, objections trending up, competitors appearing, stakeholders going silent, and messaging themes gaining or losing traction — so you know exactly what to work on today.

Usage

/octave:signals [--period <time-range>] [--focus <area>]

Options

• --period <range> - Lookback window (today, 3d, week, 2w — default: week)
• --focus <area> - Zoom into one section (deals, patterns, pipeline, content)
• --segment <name> - Filter by segment
• --motion <name> - Filter by Motion

Examples

/octave:signals                          # Full morning briefing (last 7 days)
/octave:signals --period today           # What happened since yesterday
/octave:signals --period 2w              # Broader two-week view
/octave:signals --focus deals            # Just the deals that need attention
/octave:signals --focus patterns         # Just emerging patterns

Instructions

When the user runs /octave:signals:

Step 1: Gather All Signal Data

Run these queries in parallel to gather the full picture. Use the period option to set date ranges (default: last 7 days).

A. Recent Events (activity stream)

list_events({
  eventTypes: ["CALL_TRANSCRIPT", "EMAIL_SENT", "EMAIL_REPLY_RECEIVED", "DEAL_WON", "DEAL_LOST", "DEAL_STAGE_CHANGE"],
  dateRange: { start: "<period start>", end: "<today>" },
  limit: 100
})

B. All Finding Types (signal extraction)

list_findings({
  extractionTypes: [
    "CALL_EXTERNAL_OBJECTIONS",
    "CALL_EXTERNAL_BUSINESS_PROBLEMS",
    "CALL_EXTERNAL_QUESTIONS_OR_CONFUSION_ABOUT_OFFERING",
    "CALL_EXTERNAL_COMPETITORS_TO_OUR_OFFERING",
    "CALL_INTERNAL_VALUE_PROP_PRESENTATIONS",
    "CALL_INTERNAL_PROOF_POINTS",
    "CALL_INTERNAL_USE_CASES_BROUGHT_UP",
    "EMAIL_OBJECTION",
    "EMAIL_COMPETITOR_MENTION",
    "EMAIL_VALUE_PROP",
    "EMAIL_PAIN_POINT"
  ],
  dateRange: { start: "<period start>", end: "<today>" },
  groupBy: "extractionType",
  limit: 200
})

C. Previous Period Findings (for trend comparison)

list_findings({
  extractionTypes: [
    "CALL_EXTERNAL_OBJECTIONS",
    "CALL_EXTERNAL_COMPETITORS_TO_OUR_OFFERING",
    "CALL_INTERNAL_VALUE_PROP_PRESENTATIONS",
    "CALL_INTERNAL_PROOF_POINTS"
  ],
  dateRange: { start: "<two periods ago>", end: "<period start>" },
  groupBy: "extractionType",
  limit: 200
})

D. Library Context (for gap detection)

list_all_entities({})

Step 2: Analyze and Prioritize Signals

Process the raw data into four signal categories. Prioritize by urgency — things that need action today come first.

Signal Priority Rules:

1. CRITICAL — Champion went silent, competitor entered a deal, deal moved backward, new objection not in Motion ICP / Objection entity
2. HIGH — Deal advanced (opportunity to accelerate), objection frequency spiking, stakeholder engagement dropped
3. MEDIUM — New patterns emerging, content performance shifts, pipeline health changes
4. INFO — Positive confirmations, stable trends, wins

Step 3: Present the Briefing

Use the --focus flag to show only the requested section, or show all four sections for the full briefing.

---

Full Briefing Output

See briefing-output-template.md for the full morning briefing output template.

Step 4: Handle Focus Mode

When --focus is specified, show only that section with expanded detail:

--focus deals: Show the Deals section with additional context per deal — include last 3 interactions, all stakeholders, full finding history.

--focus patterns: Show the Patterns section with full finding breakdowns — include specific quotes, event links, week-over-week trend charts.

--focus pipeline: Show Pipeline Health with deal-level detail — list every deal that moved, every deal at risk, every new deal.

--focus content: Show Content Performance with usage rankings across all proof points, value props, and Motion ICPs.

Step 5: Drill Down

When user asks to dive deeper into any signal:

get_event_detail({
  eventOId: "<event_oId>"
})

Present full context for the signal — the event, the finding, the deal history, and recommended actions.

Step 6: Act on Signals

When user wants to act on a detected gap:

Library Gap → Create/Update:

# If objection not in a Motion ICP cell, update that cell's narrative
update_motion_playbook({
  motionPlaybookOId: "<motion_playbook_oId>",
  instructions: "Add objection handling for: [detected objection] to the relevant Motion ICP narrative section"
})

# If competitor not tracked
search_knowledge_base({
  query: "<competitor name>",
  entityTypes: ["competitor"]
})
# If not found, suggest:
# /octave:library create competitor "[name]"

Deal Signal → Route to Skill: Suggest the appropriate follow-up skill based on signal type:

• Silent champion → /octave:pipeline stalled <domain>
• New competitor → /octave:battlecard --competitor "<name>"
• Deal advanced → /octave:pipeline close <domain>
• New stakeholder needed → /octave:pipeline multi-thread <domain>
• Objection spike → /octave:enablement objections --topic "<topic>"
• Win pattern → /octave:wins-losses --status won

Signal Detection Logic

Deal Signals

| Signal | Detection | Priority | |--------|-----------|----------| | Champion silent | No reply from primary contact in >2x their avg response time | CRITICAL | | New competitor | Competitor extraction type appears for a deal where it wasn't before | CRITICAL | | Deal moved backward | Deal stage change event where new stage is earlier than previous | CRITICAL | | Deal advanced | Deal stage change event moving forward | HIGH | | Stalled deal | Active deal with no events in 14+ days | HIGH | | New deal | First event for a company domain | MEDIUM | | Deal won | DEAL_WON event | INFO | | Deal lost | DEAL_LOST event | INFO |

Pattern Signals

| Signal | Detection | Priority | |--------|-----------|----------| | Objection spike | >2x increase in extraction count vs previous period | HIGH | | New objection | Objection theme appears that wasn't in previous period | HIGH | | Competitor trending | Competitor mentions increased >50% vs previous period | HIGH | | Hot proof point | Proof point cited in >3 conversations this period | MEDIUM | | Cold proof point | Proof point with 0 citations for 30+ days | MEDIUM |

Library Gap Signals

| Signal | Detection | Priority | |--------|-----------|----------| | Unaddressed objection | Objection finding with no matching Motion ICP narrative content | HIGH | | Unknown competitor | Competitor mention with no competitor entity in library | HIGH | | Missing persona | Person qualified to a persona type not in library | MEDIUM | | Stale Motion ICP | Motion ICP cell with 0 event associations in 30+ days | MEDIUM |

MCP Tools Used

Event & Finding Access

• list_events - Activity stream with date/type filters
• list_findings - Finding aggregates with grouping and trend data
• get_event_detail - Drill into specific events

Library Context

• list_all_entities - Full library inventory for gap detection
• list_motions - List Motions in the workspace
• list_motion_playbooks - List Motion Playbooks under a Motion
• list_motion_icps - List Motion ICP cells under a Motion
• find_motion_icp - Fetch Motion ICP narrative + Learning Loop learnings
• search_knowledge_base - Match findings to library content
• get_entity - Get entity details for context

Library Updates

• update_entity - Apply detected gap fixes
• create_entity - Create new entities for detected gaps
• update_motion_playbook - Edit Motion Playbook narrative sections (e.g., add objection handling to a Motion ICP)

Error Handling

No Events in Period:

No activity found in the last [period].

This could mean:

1. No calls/emails have been synced recently
2. The period is too narrow — try --period 2w or --period month
3. CRM/email integration may need reconnecting

Check your integrations in Octave, or try a wider window.

No Findings Extracted:

Events found but no findings extracted yet.

Recent events may still be processing. Check back in a few minutes. In the meantime, here's what happened (raw activity): [Show event summary without finding analysis]

Insufficient Data for Trends:

Not enough historical data to detect trends.

I'll show you what's happening now, but trend analysis needs at least 2 weeks of data. Keep using Octave and trends will appear in future briefings.

Related Skills

• /octave:insights - Deep dive into specific finding types
• /octave:pipeline - Deal-level coaching for flagged deals
• /octave:wins-losses - Pattern analysis across deal outcomes
• /octave:battlecard - Competitive intelligence for new competitors
• /octave:enablement - Turn trending objections into team materials
• /octave:audit - Full library health check
• /octave:library - Create/update entities for detected gaps