Introduction¶

When running self-media accounts, dashboards often pile up metrics like completion rate, engagement rate, GPM, etc., yet it remains hard to answer “where the problem is, what to fix first, and when changes will show.” Common practices rely on experience—tweaking titles, chasing trends—or sporadic competitor benchmarking, without weighted scoring, platform algorithm differences, or a prioritized execution checklist.

Below is the SkillHub community skill “Self-Media Account Operations Diagnosis & Optimization Expert”: given user-provided account data, it performs a systematic breakdown using an MCN operations director-level framework, outputting a five-section diagnostic report and a weekly execution schedule. The skill coordinate is @user_442551c3/self-media-diagnosis-optimizer, currently version 1.0.0, categorized under content-creation, tagged requires_api_key=false, requiring no additional API Key.

What Is This¶

One-sentence positioning: an account operations diagnosis Skill covering six major platforms—Xiaohongshu (RED), Douyin, Toutiao, WeChat Official Accounts, WeChat Channels, and Bilibili—turning issues like “slow follower growth, difficult monetization, weak content” into concrete metric gaps, root causes, and fixes.

Maintenance and source: published on SkillHub by community publisher user_442551c3 (source: community). Official entry points:

• API address: https://api.skillhub.cn/user_442551c3/self-media-diagnosis-optimizer
• Skill page: https://www.skillhub.cn/skills/user_442551c3/self-media-diagnosis-optimizer

The skill description emphasizes: it does not just conclude “engagement rate is low,” but combines traffic pool progression, title hooks, and benchmark account topics to deliver actionable improvement actions and time expectations (for strategic reference only, not promising specific follower growth or revenue).

Core Features¶

All capabilities below come from the SkillHub page and fetched descriptions, organized by usage order.

Diagnosis Framework & Report Structure¶

The skill completes systematic diagnosis across six work dimensions: diagnostic framework, data analysis, content breakdown, platform algorithms, track benchmarking, and optimization plan. The deliverable is a standard five-section diagnostic report:

1. Current status assessment
2. Problem identification
3. Track benchmarking
4. Optimization plan
5. Execution schedule

The introduction also mentions “8-dimension systematic breakdown,” appearing alongside the six-dimension workflow and five-dimension scoring model above; when using it, follow the report structure output by the skill.

Five-Dimension Diagnostic Scoring Model¶

Weighted scoring dimensions and weights:

Twelve Core Metrics Precisely Calculated¶

Metrics like completion rate, engagement rate, save rate, share rate, follower growth rate, follower acquisition cost, GPM, etc., are compared item by item against track benchmarks, quantifying gaps.

P0–P3 Problem Prioritization & Scheduling¶

After prioritization, actions are scheduled precisely by week, consistent with the “4-week execution schedule” in official examples.

Content & Platform-Specific Capabilities¶

• Viral / Low-quality determination: a four-quadrant content matrix, with item-by-item breakdowns across 6 aspects: title, cover, body text, tags, posting time, and interaction guidance.
• Algorithm differences across six platforms: e.g., Xiaohongshu emphasizes CES score and search weighting; Douyin looks at traffic pool progression metrics; Official Accounts look at open rate and follower reach; WeChat Channels look at social fission coefficients; Bilibili looks at coin-to-favorite ratios, etc. (expanded by platform within the skill).
• Twelve track benchmarking libraries: beauty, knowledge, workplace, maternal & infant, agriculture/rural/farmer (Three Rurals), food, fashion, digital/tech, film & TV, sports, finance, and relationships/emotions, with built-in industry benchmark reference ranges.

Installation & Activation¶

The skill is based on the standard SKILL.md format. Before installing, the SkillHub CLI must already be available on the machine; if not yet installed, you can first check the version via the official installation documentation:

command -v skillhub && skillhub --version

If not installed, use the recommended method from SkillHub documentation (installing both CLI and default Skills):

curl -fsSL https://skillhub-1388575217.cos.ap-guangzhou.myqcloud.com/install/install.sh | bash

To install only the CLI:

curl -fsSL https://skillhub-1388575217.cos.ap-guangzhou.myqcloud.com/install/install.sh | bash -s -- --cli-only

When installing this skill, you must use --dir to point to the current Agent’s skills directory; otherwise the default ./skills/ may not be recognized. The official example command is as follows (replace <skills directory> with the actual path):

skillhub install @user_442551c3/self-media-diagnosis-optimizer --dir <skills 目录>

Common client skills directories (from SkillHub installation docs):

• Cursor: ~/.cursor/skills/
• Claude Code: ~/.claude/skills/
• Codex: ~/.codex/skills/ or .agents/skills/ under the project

You can also ask the AI in conversation to fetch and install the skill per the installation instructions; the prompt template is:

Please install `@user_442551c3/self-media-diagnosis-optimizer` according to https://skillhub.cn/install/skillhub.md.

Full installation workflow and priority source configuration: https://skillhub.cn/install/skillhub.md

Typical Usage¶

The official reproducible example is as follows.

Input example:

Help me diagnose the Xiaohongshu account @美妆达人小A, with data from the last 30 days. The goal is follower growth and sales conversion.

Expected output highlights (as described in the skill documentation):

1. Comparison table of core metrics vs. beauty track benchmarks
2. 3 P1 severe issues + 2 P2 moderate issues, each with root cause and improvement action
3. Differentiated opportunities by benchmarking against 3 competitor accounts
4. 4-week execution schedule (what to do each week)

In actual conversations, you should also provide the account’s recent-period data (views, interactions, follower changes, sales-related metrics, etc.); the more complete the data, the less the diagnosis relies on “estimated values.”

Applicable Scenarios & Notes¶

Who It’s For¶

Precautions¶

1. No fabricated data: values not provided by the user will be marked as “estimated values.”
2. No guaranteed results: no promises on follower growth or revenue; recommendations are operational strategy references.
3. Quality depends on input: diagnostic quality depends on the completeness of user-provided data.

The SkillHub page shows this skill with a score of 100000 and 2634 downloads (statistics update with the platform).

Conclusion¶

The “Self-Media Account Operations Diagnosis & Optimization Expert” converges multi-platform account issues into weighted scoring, metric benchmarking, prioritized fixes, and weekly scheduling—suitable for operational scenarios needing deliverable diagnostic reports rather than scattered advice. For installation and details, see:

https://www.skillhub.cn/skills/user_442551c3/self-media-diagnosis-optimizer