Skill Analytics v2.0#

Install: clawhub install skill-analytics

ClawHub skill portfolio monitoring with state tracking. Remembers what it recommended.

Language#

Detect from user's message language. Default: English.

State Files#

All state stored in memory/skill-analytics/:

FilePurposestate.jsonRotation day, last run, recommendation IDsrecommendations.mdActive recommendations with statusideas-tried.mdTopics already covered (avoid repeats)

Create directory and files on first run if they don't exist.

Day-of-Week Rotation#

Use day-of-week (Monday=1, Sunday=7) from state.json (not current calendar day — track continuously):

DayFocus1Adoption Funnel2Competitive Analysis3Content & Copy4Feature Gap5Monetization6Cross-Promotion7Wildcard

After each run: increment day in state.json, wrap at 7.

Anti-Repetition Protocol#

Before generating recommendations:

1. Read memory/skill-analytics/recommendations.md
2. Read memory/skill-analytics/ideas-tried.md
3. Skip any recommendation already listed as "Pending" or already tried
4. Only generate NEW recommendations
5. If no new insights exist: say "No new recommendations this run. Previous {N} are still pending."

Recommendation Format#

markdown

| # | Recommendation | Date | Status | Result |
|---|---------------|------|--------|--------|
| 1 | **Short title** — one-line action | YYYY-MM-DD | Pending | - |

After user marks as done: change Status to "✅ Completed" with Result.

Data Collection#

Use built-in tools only (web_fetch, web_search):

text

web_fetch https://clawhub.ai/tommot2/{slug}
web_search "clawhub {skill category}"

Extract: downloads, installs, stars, version count.

No CLI tools, no npm packages, no credentials.

Output Format#

text

## 📊 Skill Analytics — {date}

### Dashboard
| Skill | DL | ⭐ | Versions |
|-------|---:|:--:|:--------:|
| ... | ... | ... | ... |

### Focus: {day focus}
{2-3 paragraphs of actual analysis. Concrete numbers.}

### New Recommendations
1. **{title}** — {one-line action}
   - Effekt: {estimated}
   - Innsats: {Lav/Middels/Høy}

### Previous Status
- Pending: {N} recommendations
- Completed: {N} recommendations
- Skipped (repeats): {N}

### Next Run
Focus: Day {N+1} — {focus}

Phase Indicator#

Based on total installs across portfolio:

PhaseInstallsFocus🌱 Seed0-10Visibility🌿 Grow10-100Conversion🌳 Scale100-1000Monetization🏢 Sustain1000+Retention

Quick Commands#

User saysAction"skill stats"Quick dashboard"skill analytics"Full analysis"fulført #3"Mark recommendation #3 as completed"alle anbefalinger"Show all with status

Guidelines for Agent#

1. Always read state before running — check recommendations and ideas-tried
2. Write state after running — update state.json, add new recommendations
3. Never repeat — check ideas-tried.md before suggesting
4. Use built-in tools only — web_fetch and web_search
5. No personal data in searches — only public ClawHub data
6. Keep output concise — max 40 lines per report
7. Language follows user

What This Skill Does NOT Do#

• Does NOT read MEMORY.md, SOUL.md, or other workspace files
• Does NOT access credentials or private data
• Does NOT use external CLI tools or npm packages
• Does NOT modify any files outside memory/skill-analytics/

More by TommoT2#

• context-brief — Persistent context survival across sessions
• setup-doctor — Diagnose and fix OpenClaw setup issues
• tommo-skill-guard — Security scanner for installed skills

Install the full suite:

bash

clawhub install skill-analytics context-brief setup-doctor tommo-skill-guard