competitive-monitoring

name	competitive-monitoring
description	Monitors competitor activity: new releases, feature announcements, pricing changes, funding rounds, acquisitions. Produces a competitive update report that feeds into the synthesis layer.
agents	{"primary":"research-agent","supporting":[]}
inputs	["discovery-config.yaml (competitors list)",".factory/discovery/competitive-baseline.md (last known state)"]
outputs	[".factory/discovery/competitive-update-YYYY-MM-DD.md",".factory/discovery/competitive-baseline.md (updated)"]

Competitive Monitoring

Monitors competitor activity continuously between discovery runs. Detects changes in the competitive landscape -- new releases, pricing shifts, funding, acquisitions, shutdowns, and new entrants -- and produces a structured competitive update report.

Trigger

• Scheduled -- weekly or as configured in discovery-config.yaml schedule.competitive_monitoring
• Manual -- human requests "Check competitor activity for [product]"
• Event-driven -- triggered by news alerts or discovery engine

Prerequisites

• discovery-config.yaml exists with products[*].competitors configured
• .factory/discovery/competitive-baseline.md exists (or will be created on first run)
• research-agent is the canonical MCP caller (DF-002); this skill spawns it for delegated research calls. Its primary tool is perplexity_research (deep research, backed by sonar-deep-research), with perplexity_search/perplexity_ask for quick lookups

Monitoring Targets

For each competitor in discovery-config.yaml:

Signal	How Detected	Urgency
New feature/release	perplexity_research: "[competitor] release notes [month year]"	HIGH if overlaps with our roadmap
Pricing change	perplexity_search: "[competitor] pricing" vs baseline	MEDIUM
Funding round	perplexity_search: "[competitor] funding"	LOW (info only)
Acquisition	perplexity_search: "[competitor] acquired"	HIGH (market shift)
Shutdown/pivot	perplexity_search: "[competitor] shutdown OR pivot"	HIGH (opportunity)
New competitor	perplexity_research: "[domain] new startup 2026"	MEDIUM

Monitoring Workflow

Step 1: Load Competitor List

Read discovery-config.yaml and extract the competitors array for the target product. If no competitors are configured, exit with no-op.

Step 2: Load Competitive Baseline

Read .factory/discovery/competitive-baseline.md for the last known state of each competitor. If the file does not exist (first run), the monitoring run establishes the baseline.

# Competitive Baseline

## Last Updated: YYYY-MM-DD

### Competitor A
- **Latest Known Release:** vX.Y.Z (YYYY-MM-DD)
- **Key Features:** [list]
- **Pricing:** [current pricing model]
- **Funding Status:** [last known round]
- **Status:** active

### Competitor B
- ...

Step 3: Research Each Competitor

For each competitor, spawn research-agent to run these queries (deep perplexity_research for release/feature scans; perplexity_search for the single-fact baseline checks):

1. Release check: "[competitor] release notes [current month] [current year]"

   • Compare against baseline's Latest Known Release
   • If new release detected: extract version, date, key features

2. Pricing check: "[competitor] pricing [current year]"

   • Compare against baseline's Pricing
   • Flag changes in pricing model, tiers, or pricing levels

3. Funding check: "[competitor] funding round [current year]"

   • Compare against baseline's Funding Status
   • Note amount, investors, valuation if available

4. Acquisition check: "[competitor] acquired OR acquisition [current year]"

   • Detect if competitor was acquired or acquired another company

5. Shutdown/pivot check: "[competitor] shutdown OR pivot OR closing [current year]"

   • Detect if competitor is shutting down or pivoting strategy

Step 4: Scan for New Entrants

Research the product's domain for new competitors (deep research — this is an open-ended landscape scan, not a single-fact lookup):

• perplexity_research: "[product domain] new startup [current year]"
• perplexity_research: "[product domain] new tool launch [current year]"
• Compare against known competitor list

Step 5: Classify Urgency

For each detected change, assign urgency:

Urgency	Criteria
HIGH	Competitor ships feature our customers are requesting; competitor acquired; competitor shuts down (opportunity)
MEDIUM	Pricing change; new entrant in our space; funding round (signals acceleration)
LOW	Minor release; informational funding news; no direct impact on our roadmap

Step 6: Produce Competitive Update

Write .factory/discovery/competitive-update-YYYY-MM-DD.md:

---
document_type: competitive-update
date: YYYY-MM-DD
product: [product-name]
competitors_monitored: [N]
changes_detected: [N]
---

# Competitive Update: YYYY-MM-DD

## Changes Detected

| Competitor | Change | Date | Urgency | Implication |
|-----------|--------|------|---------|------------|
| [name] | [change] | [date] | HIGH/MED/LOW | [what it means for us] |

## No Change (stable competitors)
[List competitors with no detected changes]

## New Entrants
[Any new competitors detected]

Step 7: Update Baseline

Update .factory/discovery/competitive-baseline.md with any newly detected state changes. Preserve history: append a ## Change Log section at the bottom of each competitor entry.

Quality Gate

• All configured competitors researched (none skipped without explanation)
• All findings dated with source URLs
• Urgency classification applied to every detected change
• Competitive baseline updated with new state

Failure Modes

• If a data source is unavailable (Perplexity timeout, API error): do not silently skip — per the research-agent's mandatory-MCP gate, flag the source as UNAVAILABLE with the verbatim error, continue with remaining sources, and note the gap in the report
• If no changes detected for any competitor: produce a "no change" report (not a silent no-op)
• If a new entrant cannot be verified: mark as UNVERIFIED and include in report with caveat

Quality Criteria

• All configured competitors researched
• Changes compared against baseline (not just raw search results)
• Urgency classification applied with rationale
• Implications assessed (not just "competitor shipped X" but "this means Y for us")
• New entrant scan covers the product domain
• Baseline updated with detected changes
• Sources cited for each finding with URLs and dates
• VERIFIED/UNVERIFIED flags on findings (same as DF-017 research quality controls)