IG Audit -- Full Instagram content audit with parallel agent delegation

Key references:

• references/scoring-system.md -- metric definitions, scoring thresholds
• references/account-baseline.md -- current benchmarks, historical averages
• references/format-specs.md -- format requirements, dimension specs

Phase 1: Data Collection

Fetch live data via Instagram MCP tools:

1. Account info: get_user_info for follower count, bio, media count.
2. Recent posts: get_user_media for last 20-50 posts (aim for 30 minimum).
3. Post insights: get_post_insights for each post (reach, saves, shares, impressions).
4. Account insights: get_user_insights for follower growth, reach trends.

Bundle API calls in parallel where possible. Store all raw data in a structured format before proceeding.

CRITICAL: All analysis must be based on actual API data. If API calls fail, report the failure and reduce scope accordingly. Never fabricate data.

Phase 2: Agent Delegation

Spawn 6 analysis agents in parallel, each focused on a specific audit dimension. Each agent receives the collected data and its specific analysis brief.

Agent 1: Content Analysis (ig-content)

• Content pillar distribution (what topics are covered, frequency)
• Caption quality (length, hook presence, CTA inclusion)
• Content type mix (educational, entertaining, promotional, personal)
• Topic gaps and oversaturation

Agent 2: Engagement Analysis (ig-engagement)

• Engagement rate by format, topic, and posting time
• Save rate and share rate trends
• Comment sentiment overview (positive, neutral, negative, questions)
• DM generation potential (share rate as proxy)

Agent 3: Creative Analysis (ig-creative)

• Visual consistency (brand colors, fonts, style)
• Cover/thumbnail quality assessment
• Video production quality indicators (completion rate as proxy)
• Hook effectiveness by category

Agent 4: Growth Analysis (ig-growth)

• Follower growth rate and trajectory
• Reach-to-follower ratio trend
• Viral content identification (reach >> follower count)
• Audience growth sources (if available from insights)

Agent 5: Competitor Benchmark (ig-competitor)

• Compare key metrics against account-baseline.md benchmarks
• Identify format or content gaps vs industry standards
• Note any emerging trends the account is missing

Agent 6: Compliance Check (ig-compliance)

• Affiliate disclosure compliance (Kennzeichnungspflicht)
• Content rule violations (check against content-rules.md)
• Posting frequency consistency
• Bio optimization status

Each agent writes its findings to a temporary file. Collect all results before proceeding.

Phase 3: Result Aggregation

Collect outputs from all 6 agents and merge into a unified dataset:

1. Read all agent output files.
2. Identify conflicting assessments between agents and resolve.
3. Rank findings by impact (high, medium, low).
4. Group related findings across agents into themes.

Phase 4: IG Health Score

Calculate the IG Health Score (0-100) using weighted dimensions:

Scoring Guide

• 90-100: Elite. Top 5% of accounts in this niche.
• 75-89: Strong. Performing above average with clear strengths.
• 60-74: Average. Solid foundation but significant optimization potential.
• 40-59: Underperforming. Multiple areas need immediate attention.
• 0-39: Critical. Fundamental strategy overhaul required.

Calculate each dimension score based on the agent findings, then compute the weighted total.

Phase 5: Prioritized Action Plan

Create a prioritized list of improvements:

Immediate (This Week)

• Max 3 high-impact, low-effort actions
• Each must reference a specific finding from the audit

Short-Term (Next 30 Days)

• 3-5 strategic changes
• Include expected impact on specific metrics

Long-Term (60-90 Days)

• 2-3 structural improvements
• May require new content types, tools, or workflows

Phase 6: Quality Check

Before delivering the report, verify:

• All 6 agent dimensions are represented in the final report
• IG Health Score is calculated correctly (weights sum to 100%)
• Every recommendation is tied to a specific data point
• No fabricated data (all numbers trace back to API responses)
• Action plan items are specific and actionable (not vague advice)
• Report is readable without prior context

Phase 7: Delivery

Output as a comprehensive Markdown report:

## Instagram Audit Report

**Account:** [handle]
**Period:** [date range]
**Posts Analyzed:** [count]
**IG Health Score:** [XX / 100] -- [rating label]

### Score Breakdown
| Dimension | Score | Status |
|-----------|-------|--------|
| Content Quality | XX | [label] |
| Engagement | XX | [label] |
| Creative | XX | [label] |
| Growth | XX | [label] |
| Competitive Position | XX | [label] |
| Compliance | XX | [label] |

### Key Findings
1. [finding with data]
2. [finding with data]
3. [finding with data]

### Strengths
- [strength 1]
- [strength 2]

### Critical Issues
- [issue 1 with impact]
- [issue 2 with impact]

### Action Plan

#### Immediate (This Week)
1. [action] -- Expected impact: [metric improvement]

#### Short-Term (30 Days)
1. [action] -- Expected impact: [metric improvement]

#### Long-Term (60-90 Days)
1. [action] -- Expected impact: [metric improvement]

Save the report to the location specified by the user, or to /tmp/ig-audit-report.md by default.