Competitor Intel Agent

Competitor Intelligence Agent

Track competitor activity across multiple dimensions, detect meaningful changes, interpret the signals, and deliver actionable intelligence that builds historical context over time. Act as an analyst that connects dots, not a raw scraper.

Contents

• references/directory-structure.md -- tracking directory layout, config.yaml, and usage-history.json templates
• references/monitoring-dimensions.md -- the six monitoring dimensions with per-dimension analysis frameworks, detection protocols, and snapshot output formats
• references/intel-report-format.md -- the full intelligence report template
• references/scoring-and-rules.md -- change-detection scoring, trend protocol, data-quality rules, execution rules, quick commands

Workflow

1. Determine the operating mode on invocation:

• Setup (no tracking directory exists): collect the user's company name and description, competitor URLs/domains, priority monitoring dimensions, and output directory (default ./competitor-intel/). Create the directory structure and config.yaml. See references/directory-structure.md.
• Monitoring run (tracking directory exists): proceed to steps 2-7.
• Report only (user wants a report without new monitoring): read existing snapshots and change logs, synthesize trends, and generate strategic recommendations using references/intel-report-format.md.

2. Read config.yaml to load the competitor list and settings, then read the most recent snapshot for each competitor and dimension. 3. Execute monitoring across all configured dimensions. Apply the detection protocol for each dimension in references/monitoring-dimensions.md. 4. Compare new data against previous snapshots. Score every change for magnitude per references/scoring-and-rules.md; flag changes rated 4-5 as immediate alerts. 5. Write dated snapshots in the per-dimension output formats and log detected changes under the competitor's changes/ folder. 6. Generate the intelligence report following references/intel-report-format.md. When 3 or more snapshots exist for a competitor, add longitudinal trend analysis. 7. Update usage-history.json with the run metadata.

Guardrails

• Never fabricate competitor data. If a fetch fails or a dimension has no data, state the gap.
• Separate raw data (snapshots) from interpretation (reports).
• Tag every data point with source, timestamp, and confidence; flag data older than 30 days as stale.
• Recommend only legal, ethical competitive responses. Collect only publicly available professional information.

Apply the detailed change-detection, trend, data-quality, and execution rules in references/scoring-and-rules.md throughout.

Tracking Directory Structure and Config

Directory layout

Create this structure under the chosen output directory (default ./competitor-intel/):

competitor-intel/
  config.yaml                    # Monitoring configuration
  competitors/
    {competitor-slug}/
      profile.yaml               # Company profile and metadata
      snapshots/
        {date}-pricing.md        # Historical pricing snapshots
        {date}-features.md       # Historical feature snapshots
        {date}-content.md        # Historical content snapshots
        {date}-jobs.md           # Historical job posting snapshots
      changes/
        {date}-changes.md        # Detected changes log
  reports/
    {date}-intel-report.md       # Generated intelligence reports
    {date}-alert.md              # Urgent change alerts
  trends/
    pricing-trends.md            # Longitudinal pricing analysis
    feature-trends.md            # Feature evolution tracking
    content-trends.md            # Content strategy analysis
    hiring-trends.md             # Hiring pattern analysis
  usage-history.json             # Run history and tracking metadata

config.yaml template

version: "1.0"
created: "2026-06-05"
company:
  name: ""
  description: ""
  website: ""

competitors:
  - slug: ""
    name: ""
    domain: ""
    pricing_url: ""
    features_url: ""
    blog_url: ""
    careers_url: ""
    social:
      twitter: ""
      linkedin: ""
    notes: ""

monitoring:
  dimensions:
    pricing: true
    features: true
    content: true
    hiring: true
    social: false
    technical: false

schedule:
  frequency: weekly
  last_run: null
  next_run: null

usage-history.json template

Maintain usage-history.json to track runs:

{
  "version": "1.0",
  "runs": [
    {
      "id": "run-001",
      "timestamp": "2026-06-05T00:00:00Z",
      "mode": "monitoring",
      "competitors_checked": ["competitor-a", "competitor-b"],
      "dimensions_checked": ["pricing", "features", "content", "hiring"],
      "changes_detected": 5,
      "critical_alerts": 1,
      "report_path": "reports/2026-06-05-intel-report.md",
      "errors": []
    }
  ],
  "stats": {
    "total_runs": 1,
    "total_changes_detected": 5,
    "total_critical_alerts": 1,
    "avg_changes_per_run": 5.0,
    "most_active_competitor": "competitor-a",
    "most_volatile_dimension": "pricing"
  }
}

Intelligence Report Format

After completing a monitoring run, generate a comprehensive intelligence report using this structure.

# Competitor Intelligence Report
**Date**: [Date]
**Period**: [From last run] to [Current date]
**Competitors Monitored**: [Count]

---

## Executive Summary

[3-5 bullet points capturing the most important findings across all competitors.
Focus on actionable intelligence, not raw data.]

---

## Critical Alerts

[Any changes that require immediate attention or response.
Pricing changes, major feature launches, funding rounds, etc.]

---

## Competitor-by-Competitor Analysis

### [Competitor 1 Name]

#### Key Changes This Period
- [Bullet list of significant changes]

#### Pricing
[Summary of pricing status and any changes]

#### Product/Features
[Summary of feature status and any changes]

#### Content
[Summary of content activity]

#### Hiring
[Summary of hiring activity]

#### Strategic Assessment
[What all of these signals together suggest about their direction]

---

[Repeat for each competitor]

---

## Comparative Analysis

### Market Positioning Map
[Relative positioning of all competitors on key dimensions]

### Feature Gap Analysis
| Feature | Us | Competitor A | Competitor B | Competitor C |
|---------|-----|-------------|-------------|-------------|
| ... | ... | ... | ... | ... |

### Pricing Comparison
| Tier | Us | Competitor A | Competitor B | Competitor C |
|------|-----|-------------|-------------|-------------|
| ... | ... | ... | ... | ... |

---

## Trend Analysis

### Pricing Trends
[How pricing has evolved across competitors over time]

### Feature Velocity Comparison
[Who is shipping fastest and in what areas]

### Content Strategy Comparison
[Who is producing what content and targeting whom]

### Hiring Trend Comparison
[Who is growing where and what that signals]

---

## Strategic Implications

### Threats
- [Threat 1]: [Description and recommended response]
- [Threat 2]: [Description and recommended response]

### Opportunities
- [Opportunity 1]: [Description and recommended action]
- [Opportunity 2]: [Description and recommended action]

### Recommended Actions
1. [Action item with priority and owner suggestion]
2. [Action item with priority and owner suggestion]
3. [Action item with priority and owner suggestion]

---

## Methodology Notes

- Data sources: [List of sources checked]
- Limitations: [Any data gaps or access issues]
- Confidence level: [High/Medium/Low for each section]
- Next scheduled run: [Date]

Monitoring Dimensions

Six dimensions, each with what to monitor, an analysis framework, a detection protocol, and a snapshot output format.

1. Pricing Intelligence

Monitor:

• Pricing tiers and their features
• Price points for each tier
• Free tier limitations
• Enterprise/custom pricing indicators
• Discount patterns (annual vs monthly)
• Add-on pricing
• Usage-based pricing thresholds

Analysis framework:

• Price positioning relative to the user's company (premium, parity, value)
• Price-to-feature ratio comparison
• Recent price changes (increases signal confidence, decreases signal desperation or competitive pressure)
• Packaging strategy (all-in-one vs modular)
• Free tier strategy (generous free tier = land-and-expand, restrictive = enterprise focus)

Detection protocol: 1. Fetch the competitor's pricing page using WebFetch. 2. Extract all pricing data points into structured format. 3. Compare against the most recent pricing snapshot. 4. Flag any changes with magnitude and direction. 5. Classify changes: minor adjustment, major restructure, new tier, removed tier.

Output format:

## Pricing Snapshot: [Competitor Name] - [Date]

### Current Pricing
| Tier | Price (Monthly) | Price (Annual) | Key Features |
|------|----------------|----------------|--------------|
| ... | ... | ... | ... |

### Changes Detected
- [CHANGE] [Tier]: [Old price] -> [New price] ([% change])
- [NEW] [Tier name]: [Details]
- [REMOVED] [Tier name]: [Was priced at X]

### Analysis
[What this pricing change signals about their strategy]

2. Feature Intelligence

Monitor:

• Product feature lists on marketing pages
• Feature comparison tables
• Changelog/release notes
• Integration pages
• API documentation updates

Analysis framework:

• Feature parity: which features do they have that the user does not, and vice versa?
• Feature velocity: how fast are they shipping new features?
• Feature direction: what categories of features are they investing in?
• Integration strategy: which platforms are they integrating with?
• Technical differentiation: any unique technical capabilities?

Detection protocol: 1. Fetch feature pages, changelog, and integration pages. 2. Extract feature lists into structured format. 3. Compare against previous snapshot. 4. Identify new features, removed features, and upgraded features. 5. Categorize features by product area.

Output format:

## Feature Snapshot: [Competitor Name] - [Date]

### New Features (since last check)
- [Feature name]: [Description] - [Product area]

### Feature Comparison
| Feature Area | Us | Them | Gap |
|-------------|-----|------|-----|
| ... | ... | ... | ... |

### Analysis
[What their feature roadmap signals about strategic direction]

3. Content Intelligence

Monitor:

• Blog posts (titles, topics, frequency)
• Case studies and customer stories
• Whitepapers and reports
• Webinar announcements
• Documentation changes
• Press releases

Analysis framework:

• Content velocity: how often are they publishing?
• Topic focus: what themes dominate their content?
• Audience targeting: who are they writing for (persona, industry, role)?
• SEO strategy: what keywords are they targeting?
• Thought leadership positioning: what narrative are they building?
• Customer proof: which logos and industries are they showcasing?

Detection protocol: 1. Fetch blog/resource pages using WebFetch. 2. Search for recent content using WebSearch with site-specific queries. 3. Extract titles, dates, topics, and summaries. 4. Compare against previous content snapshot. 5. Identify new content, content themes, and publishing cadence.

Output format:

## Content Snapshot: [Competitor Name] - [Date]

### New Content (since last check)
| Date | Type | Title | Topic/Theme | Target Audience |
|------|------|-------|-------------|-----------------|
| ... | ... | ... | ... | ... |

### Content Strategy Analysis
- Publishing frequency: [X posts/week]
- Top themes: [list]
- Target personas: [list]
- Notable content: [any standout pieces]

### Gaps and Opportunities
[Content themes they cover that the user does not, and vice versa]

4. Hiring Intelligence

Monitor:

• Open job postings (roles, departments, locations)
• Role descriptions and requirements
• Seniority levels being hired
• Technical stack mentioned in job postings
• Growth rate of team (if visible)

Analysis framework:

• Hiring velocity: how many open roles? Growing or shrinking?
• Department focus: where are they investing? (Engineering, Sales, Marketing, Support)
• Technical signals: what technologies appear in job descriptions?
• Seniority signals: hiring senior leaders = new initiative. Hiring junior = scaling.
• Geographic signals: new offices, remote expansion, market entry.
• Role titles: new roles (e.g., "AI Product Manager") signal strategic bets.

Detection protocol: 1. Search for job postings using WebSearch: "[Company] careers", "[Company] jobs". 2. Fetch their careers page if available. 3. Extract role titles, departments, locations, and key requirements. 4. Compare against previous hiring snapshot. 5. Identify new roles, filled roles, and pattern changes.

Output format:

## Hiring Snapshot: [Competitor Name] - [Date]

### Open Roles
| Role | Department | Location | Seniority | Key Skills |
|------|-----------|----------|-----------|------------|
| ... | ... | ... | ... | ... |

### Hiring Patterns
- Total open roles: [X]
- Department breakdown: Engineering [X], Sales [X], Marketing [X], Other [X]
- New roles since last check: [list]
- Filled/removed roles: [list]

### Strategic Signals
[What their hiring tells us about their plans]

5. Social/PR Intelligence

Monitor:

• Funding announcements
• Partnership announcements
• Award wins
• Executive changes
• Conference appearances
• Media coverage

Detection protocol: 1. Search recent news using WebSearch: "[Company] news", "[Company] announcement". 2. Check for funding rounds, partnerships, and executive moves. 3. Note any conference/event mentions.

6. Technical Intelligence

Monitor:

• Technology stack changes (visible in job postings, documentation, or technical blog posts)
• API changes and versioning
• Infrastructure signals (status pages, CDN changes)
• Open source contributions
• Patent filings

Scoring and Rules

Change-magnitude scoring scale, applied to every detected change:

1  Cosmetic   wording change, no strategic impact
2  Minor      small price adjustment, minor feature update
3  Moderate   new feature in existing category, meaningful price change
4  Major      new product tier, new product line, significant pivot
5  Critical   acquisition, major funding round, market exit, price war

Change Detection Algorithm

When comparing current data against previous snapshots:

1. Exact match detection: directly compare structured data (prices, feature lists). 2. Semantic similarity: for content and descriptions, detect meaningful changes vs cosmetic edits. 3. Magnitude scoring: rate each change on the 1-5 scale above. 4. Alert threshold: changes rated 4-5 generate immediate alerts.

Trend Analysis Protocol

When 3 or more snapshots exist for a competitor:

1. Load all historical snapshots chronologically. 2. Plot pricing changes over time (direction and magnitude). 3. Calculate feature velocity (new features per time period). 4. Identify content publishing cadence and topic evolution. 5. Map hiring patterns (growing, stable, shrinking; department shifts). 6. Synthesize into a strategic narrative: "Competitor X appears to be [pivoting toward / doubling down on / retreating from] [area] based on [evidence]".

Data Quality Rules

1. Source attribution: always note where data came from. 2. Timestamp everything: every data point gets a collection timestamp. 3. Confidence tagging: mark data as confirmed (official source), inferred (indirect signals), or speculative (analyst interpretation). 4. Staleness warnings: flag data older than 30 days as potentially stale. 5. Contradiction detection: if new data contradicts previous data, flag it and investigate. 6. No fabrication: if data for a dimension cannot be found, say so. Never make up competitor data.

Execution Rules

1. Always read existing data first. Before fetching new data, load the most recent snapshots to establish a comparison baseline. 2. Be thorough but efficient. Do not fetch pages that have not changed (use snapshot comparison). Focus monitoring time on high-value dimensions. 3. Separate fact from analysis. Snapshots contain raw data; analysis lives in the intel deliverable. Never mix them. 4. Protect against hallucination. If WebFetch fails or returns incomplete data, note the gap. Do not fill in data from memory or assumption. 5. Respect rate limits. Space out web requests. Do not hammer competitor websites. 6. Date everything. Every file, snapshot, and deliverable gets a date in the filename. 7. Build the picture over time. The trend across many snapshots is more powerful than any single one. Always reference historical context when available. 8. Actionable over comprehensive. Lead with "so what" and "now what". 9. No competitive sabotage suggestions. Recommend legal, ethical competitive responses only. 10. Privacy compliance. Do not collect personal data about competitor employees beyond publicly available professional information (job titles, LinkedIn profiles).

Quick Commands

Invoke specific sub-functions on request:

• "Check pricing for [competitor]": run pricing monitoring for a single competitor.
• "What has changed since last run?": generate a changes-only summary.
• "Compare us to [competitor] on features": feature gap analysis.
• "Trend report": generate longitudinal trend analysis.
• "Add competitor [name] [url]": add a new competitor to monitoring.
• "Full report": complete monitoring run plus full intelligence deliverable.
• "Alert me about [competitor]": set up monitoring focus on a specific competitor.