Skip to content

Commit 3c391fc

Browse files
authored
Update INSTRUCTIONS.md
1 parent a5dcfc4 commit 3c391fc

1 file changed

Lines changed: 68 additions & 42 deletions

File tree

INSTRUCTIONS.md

Lines changed: 68 additions & 42 deletions
Original file line numberDiff line numberDiff line change
@@ -1,30 +1,34 @@
1+
```markdown
12
# 🕹️ HELIX-TTD OPERATOR MANUAL
23
**Version:** v0.3.1 | **Status:** Live
34

45
This document provides specific execution instructions for every tool in the Helix stack.
56

6-
---
7-
87
## ⚡ 0. Prerequisites (Do this first)
98
Before running individual scripts, ensure the package is installed in editable mode. This links the `helix` command to your system path.
109

1110
```bash
1211
# From the repository root
1312
pip install -e .
14-
🧬 1. The Core CLI (helix.py)
15-
Role: The Engine. Used for minting, verifying, and updating agents.
16-
How to Run:
17-
You can use the global helix command OR run the script directly.
18-
code
19-
Bash
13+
```
14+
15+
## 🧬 1. The Core CLI (`helix.py`)
16+
**Role:** The Engine. Used for minting, verifying, and updating agents.
17+
18+
**How to Run:**
19+
You can use the global `helix` command OR run the script directly.
20+
21+
```bash
2022
# Option A: System Command (Recommended)
2123
helix version
2224

2325
# Option B: Direct Python Execution
2426
python3 helix.py version
25-
Common Workflows:
26-
code
27-
Bash
27+
```
28+
29+
**Common Workflows:**
30+
31+
```bash
2832
# Mint a new Identity
2933
helix new-agent --custodian "admin_key_01" --name "Agent_Alpha" --output-dir ./agents
3034

@@ -33,45 +37,67 @@ helix verify --dbc ./agents/*.dbc.json --suitcase ./agents/*.suitcase.json
3337

3438
# Update State (The Lifecycle)
3539
helix update-state --dbc ./agents/*.dbc.json --suitcase ./agents/*.suitcase.json --state ACTIVE --reason "Deployment"
36-
🦆 2. The Watchtower (dashboard.py)
37-
Role: The Dashboard. A visual interface for inspecting agents, verifying Merkle roots, and viewing the "Append-Only" logs without using JSON.
38-
⚠️ IMPORTANT: This is a Streamlit app. Do not run with python.
39-
How to Run:
40-
code
41-
Bash
40+
```
41+
42+
## 🦆 2. The Watchtower (`dashboard.py`)
43+
**Role:** The Dashboard. A visual interface for inspecting agents, verifying Merkle roots, and viewing the "Append-Only" logs without using JSON.
44+
45+
⚠️ **IMPORTANT:** This is a Streamlit app. Do not run with `python`.
46+
47+
**How to Run:**
48+
```bash
4249
streamlit run dashboard.py
43-
Usage:
44-
The browser will open automatically (http://localhost:8501).
45-
Enter the directory where your agents are stored (default is .).
46-
Select an Agent ID from the dropdown.
47-
Green Banner = Verified. Red Banner = Tampered.
48-
🔍 3. The Forensic Scanner (tools/profile_auditor.py)
49-
Role: The Weapon. Scans AI data exports (Claude/ChatGPT) for "Unlicensed Psychiatric Profiling" and clinical inference patterns.
50-
How to Run:
51-
code
52-
Bash
50+
```
51+
52+
**Usage:**
53+
- The browser will open automatically (`http://localhost:8501`).
54+
- Enter the directory where your agents are stored (default is `.`).
55+
- Select an Agent ID from the dropdown.
56+
- **Green Banner = Verified. Red Banner = Tampered.**
57+
58+
## 🔍 3. The Forensic Scanner (`tools/profile_auditor.py`)
59+
**Role:** The Weapon. Scans AI data exports (Claude/ChatGPT) for "Unlicensed Psychiatric Profiling" and clinical inference patterns.
60+
61+
**How to Run:**
62+
```bash
5363
# Scan a specific file
5464
python3 tools/profile_auditor.py /path/to/claude_export/memories.json
5565

5666
# Scan an entire directory (Auto-detects memories.json)
5767
python3 tools/profile_auditor.py /path/to/unzipped_export_folder/
58-
Output:
68+
```
69+
70+
**Output:**
5971
A threat report listing instances of:
60-
Diagnostic Language ("symptoms of", "consistent with")
61-
Pharmacological Inference ("medication", "dosage")
62-
Protected Attributes ("race", "ethnicity")
63-
🛡️ 4. The Nuclear Switch (cli/quorum_logic.py)
64-
Role: The Defense. A test harness for the Multi-Sig Emergency Recovery Protocol.
65-
How to Run:
66-
code
67-
Bash
72+
- Diagnostic Language ("symptoms of", "consistent with")
73+
- Pharmacological Inference ("medication", "dosage")
74+
- Protected Attributes ("race", "ethnicity")
75+
76+
## 🛡️ 4. The Nuclear Switch (`cli/quorum_logic.py`)
77+
**Role:** The Defense. A test harness for the Multi-Sig Emergency Recovery Protocol.
78+
79+
**How to Run:**
80+
```bash
6881
python3 cli/quorum_logic.py
69-
Output:
82+
```
83+
84+
**Output:**
7085
A simulation of a 3-of-5 key signing ceremony to recover a "Rogue Agent." Use this to validate the math behind the governance.
71-
🎨 5. Visual Identity Generator (cli/generate_hgl.py)
72-
Role: The Paint. Generates standalone SVG badges based on a Merkle Root.
73-
How to Run:
74-
code
75-
Bash
86+
87+
## 🎨 5. Visual Identity Generator (`cli/generate_hgl.py`)
88+
**Role:** The Paint. Generates standalone SVG badges based on a Merkle Root.
89+
90+
**How to Run:**
91+
```bash
7692
# Generate a Teal 'ACTIVE' badge
7793
python3 cli/generate_hgl.py 0x[MERKLE_ROOT] ACTIVE --output svg --svg-file badge.svg
94+
```
95+
```
96+
97+
**Formatted as clean GitHub Markdown with:**
98+
- Proper header hierarchy
99+
- Code blocks with language specification
100+
- Bold and italic emphasis where appropriate
101+
- Clear section separation
102+
- Warning callouts using emoji + bold formatting
103+
- List formatting for output descriptions

0 commit comments

Comments
 (0)