Proof EngineSAFE

From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.

python3 /workspace/proof/scripts/proof_manager.py init
python3 /workspace/proof/scripts/proof_manager.py status

SKILL.md                          ← Skill instructions
README.md                         ← This file
scripts/proof_manager.py          ← CLI: all engines
templates/dashboard.json          ← Financial dashboard template
templates/proof_item.json         ← Proof item template
templates/opportunity_template.json ← Opportunity scoring template
references/storytelling.md        ← Story arc frameworks
.learnings/LEARNINGS.md           ← Auto-populated
AUDIT.md                          ← Auto-populated

The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.

---
name: proof-engine
description: >
  Transforms every result [PRINCIPAL_NAME] achieves into deployable proof
  across all business domains. Captures P&L, agent performance, funnel
  revenue, testimonials, milestones, and media mentions. Converts raw data
  into compelling stories via the Storytelling Engine. Generates
  proof-based content ready for all platforms. Tracks a multi-channel
  financial dashboard. Scans high-potential business opportunities for
  2026. Deploys proof automatically into funnels, brand, outreach, and
  VSL scripts. The credibility engine that makes everything else convert.
version: 1.1.0
author: Wesley Armando (Georges Andronescu)
license: MIT
metadata:
  openclaw:
    emoji: "💎"
    security_level: L2
    required_paths:
      read:
        - /workspace/proof/
        - /workspace/brand/
        - /workspace/CASHFLOW/
        - /workspace/voice/
        - /workspace/memory/
        - /workspace/revenue/
        - /workspace/content/
        - /workspace/.learnings/
      write:
        - /workspace/proof/vault/
        - /workspace/proof/dashboard.json
        - /workspace/proof/stories/
        - /workspace/proof/opportunities/
        - /workspace/proof/content/
        - /workspace/proof/AUDIT.md
    network_behavior:
      makes_requests: true
      request_targets:
        - https://api.telegram.org (Telegram Bot API — requires TELEGRAM_BOT_TOKEN)
      uses_agent_telegram: true
    requires.env:
      - TELEGRAM_BOT_TOKEN
      - TELEGRAM_CHAT_ID
---

# Proof Engine — Credibility at Scale

> "Nobody believes what you say. Everyone believes what you prove."

This skill turns every result [PRINCIPAL_NAME] achieves — in any business,
on any platform — into automated credibility that feeds every other skill.

```
ENGINE 1 — CAPTURE
  Collects proof from all business domains automatically
  Sources: trading, agents, funnels, content, clients, media

ENGINE 2 — VAULT
  Structured storage with impact scoring
  Multi-channel financial dashboard — all revenue in one view

ENGINE 3 — STORYTELLING
  Transforms raw data into compelling narratives
  Before → Trigger → Result → Mission arcs ready to deploy

ENGINE 4 — CONTENT PROOF
  Generates proof-based content for all platforms
  Real numbers, real results — zero hype

ENGINE 5 — OPPORTUNITY
  Scans high-potential business opportunities for 2026
  Filter: digital, automatable, scalable, high-margin

ENGINE 6 — DEPLOY
  Injects proof at the right moment into every other skill
  Funnels, brand, outreach, VSL, email sequences
```

---

## ENGINE 1 — CAPTURE

### Proof Sources — All Business Domains

```
DOMAIN 1 — TRADING & CRYPTO
  Source:    /workspace/CASHFLOW/ (crypto-executor output)
  Captures:  Monthly P&L, win rate, max drawdown, best trade
  Format:    { domain, date, metric, value, context, impact_score }
  Schedule:  Daily auto-capture at 06h00

DOMAIN 2 — AI AGENTS & AUTOMATION
  Source:    /workspace/ agent logs, AUDIT.md files
  Captures:  Tasks automated, time saved, errors resolved autonomously,
             skills deployed, uptime percentage
  Format:    { domain, date, task, outcome, hours_saved, impact_score }
  Schedule:  Weekly auto-capture Monday 08h00

DOMAIN 3 — FUNNELS & DIGITAL SALES
  Source:    /workspace/CASHFLOW/ funnel revenue data
  Captures:  Revenue per funnel, conversion rates, leads generated,
             best-performing offer, email open rates
  Format:    { domain, date, funnel, revenue, conversions, impact_score }
  Schedule:  Weekly auto-capture Monday 08h00

DOMAIN 4 — CONTENT & AUDIENCE
  Source:    /workspace/brand/AUDIT.md
  Captures:  Follower milestones, viral posts, engagement rates,
             platform growth (Twitter, LinkedIn, Instagram, YouTube, TikTok)
  Format:    { domain, platform, date, metric, value, impact_score }
  Schedule:  Weekly auto-capture

DOMAIN 5 — CLIENTS & TESTIMONIALS
  Source:    Manual entry + /workspace/proof/vault/testimonials/
  Captures:  Written testimonials, video testimonials, client results,
             before/after transformations
  Format:    { domain, date, client, result, quote, impact_score }
  Schedule:  On-demand (manual trigger)

DOMAIN 6 — PRODUCTS & DIGITAL ASSETS
  Source:    Manual entry + revenue data
  Captures:  Product sales, downloads, reviews, refund rate,
             revenue per product
  Format:    { domain, date, product, sales, revenue, reviews }
  Schedule:  Monthly auto-capture

DOMAIN 7 — MEDIA & AUTHORITY
  Source:    Manual entry
  Captures:  Podcast appearances, press mentions, newsletter features,
             speaking invitations, partnership deals
  Format:    { domain, date, source, type, reach, link }
  Schedule:  On-demand (manual trigger)

DOMAIN 8 — PARTNERSHIPS & COLLABS
  Source:    Manual entry
  Captures:  Deals signed, revenue generated, reach amplified,
             affiliate commissions
  Format:    { domain, date, partner, type, outcome, revenue }
  Schedule:  On-demand (manual trigger)
```

### Impact Scoring System

```
IMPACT SCORE (1-10) — determines deployment priority

Score 9-10 : HERO PROOF — deploy everywhere immediately
  Examples: First €10K month, 100K followers, viral post 1M+ views

Score 7-8  : STRONG PROOF — deploy in funnels + brand
  Examples: Consistent €3K/month, testimonial with specific results

Score 5-6  : SUPPORTING PROOF — deploy in content + outreach
  Examples: New tool working well, 10% conversion rate

Score 3-4  : CONTEXT PROOF — use in nurture sequences
  Examples: Learning moment, process improvement, milestone progress

Score 1-2  : ARCHIVE — store but don't deploy actively
  Examples: Small wins, early-stage data
```

### Auto-Capture CLI

```bash
# Capture from all domains
python3 /workspace/proof/scripts/proof_manager.py capture --all

# Capture specific domain
python3 /workspace/proof/scripts/proof_manager.py capture \
  --domain trading

# Manual entry
python3 /workspace/proof/scripts/proof_manager.py add \
  --domain clients \
  --metric testimonial

SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.

What the source does

Filesystem

none-observed

Network

declared (3 observation(s))

Shell

none-observed

Dependencies

pinned

Secrets in source

none-found

Findings (1)

<key>k5Ey9KFlkqpj+SDkUw+5ED9lTA3En/qUi0zdrydUCH3kMWTE3Eh65NXnFCaxlY2omY2JHnlEoK7Li7oOEvM7eG5VPdcO/sFlMfoCRdnLYdepJ+uLzYwOWR8W4yQVve/clxVFTVRL4DFleKInGdpAxIbHZT2yi4ADAMENls1N1XSLojRuqXePXDeAT/4Mv4TTx0s

Gates applied: no_behavioural_pass.