charlieviettq/awesome-agent-skill / .claude/skills/algo-social-influence
"algo-social-influence"
Verified"Measure social media influence using engagement-weighted metrics beyond follower count. Use this skill when the user needs to evaluate influencer effectiveness, compare influence across accounts, or build an influence scoring system — even if they say 'who is more influential', 'influencer ranking', or 'measure social impact'.".
Social Influence Measurement
Overview
Influence scoring evaluates an account's ability to drive actions (engagement, sharing, conversions) beyond mere reach. Combines reach, resonance (engagement depth), and relevance (topical authority). Computes as weighted composite score.
When to Use
Trigger conditions:
- Evaluating and comparing influencers for marketing campaigns
- Building an influence scoring or ranking system
- Assessing brand ambassador effectiveness
When NOT to use:
- When measuring content virality dynamics (use viral spread models)
- When computing basic engagement rates (use engagement rate calculator)
Algorithm
IRON LAW: Follower Count ≠ Influence
Influence requires ENGAGEMENT. An account with 1M followers and
0.01% engagement rate has less influence than one with 10K followers
and 5% engagement. Measure: reach × engagement rate × relevance.
Phase 1: Input Validation
Collect per account: follower count, avg likes/comments/shares per post, posting frequency, audience demographics, topic categories. Gate: Minimum 20 recent posts for stable metrics.
Phase 2: Core Algorithm
- Reach score: Normalize follower count to log scale (diminishing returns)
- Engagement score: (avg engagements / followers) × 100, weighted by type (share > comment > like)
- Relevance score: Topic overlap between influencer content and target campaign
- Composite: Influence = w₁×Reach + w₂×Engagement + w₃×Relevance (weights tuned per campaign goal)
- Adjust for: audience authenticity (bot follower %), post frequency consistency
Phase 3: Verification
Spot-check: do high-scoring accounts actually drive actions? Cross-reference with historical campaign performance data if available. Gate: Top-ranked accounts have demonstrable engagement history.
Phase 4: Output
Return ranked influence scores with component breakdown.
Output Format
{
"rankings": [{"account": "@handle", "influence_score": 82, "reach": 75, "engagement": 90, "relevance": 85}],
"metadata": {"accounts_analyzed": 50, "weights": {"reach": 0.2, "engagement": 0.5, "relevance": 0.3}}
}
Examples
Sample I/O
Input: Account A: 500K followers, 0.5% engagement. Account B: 50K followers, 4.2% engagement. Same relevance. Expected: B scores higher due to engagement dominance in weighting.
Edge Cases
| Input | Expected | Why |
|---|---|---|
| Viral one-hit account | High recent engagement, low stability | Need temporal consistency check |
| Celebrity with low engagement | High reach, low influence per dollar | Reach-only strategy, expensive |
| Micro-influencer niche | High relevance + engagement | Best ROI for targeted campaigns |
Gotchas
- Fake engagement: Bot likes/comments inflate metrics. Use authenticity tools (HypeAuditor, etc.) to detect.
- Platform differences: 2% engagement on Instagram is average; 2% on Twitter/X is excellent. Normalize by platform benchmarks.
- Engagement pods: Groups of influencers artificially engaging with each other's content. Check if engagement comes from diverse sources.
- Influence ≠ conversion: High engagement doesn't guarantee purchase intent. Track downstream metrics (link clicks, promo code usage) for campaign ROI.
- Temporal decay: Influence changes. Quarterly reassessment is minimum; monthly is better for fast-moving categories.
References
- For audience authenticity detection methods, see
references/authenticity-detection.md - For influencer ROI measurement framework, see
references/influencer-roi.md
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |