50c 3.9.3 → 3.9.4
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +201 -175
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -1,175 +1,201 @@
|
|
|
1
|
-
# 50c
|
|
2
|
-
|
|
3
|
-
AI developer tools via MCP. Pay-per-use from $0.01. No subscriptions.
|
|
4
|
-
|
|
5
|
-
**Requirements:** Node.js 18+
|
|
6
|
-
|
|
7
|
-
## Quick Start
|
|
8
|
-
|
|
9
|
-
```bash
|
|
10
|
-
# Install and configure for your IDE
|
|
11
|
-
npx 50c install
|
|
12
|
-
|
|
13
|
-
# Set API key (get one at https://50c.ai)
|
|
14
|
-
50c config key YOUR_API_KEY
|
|
15
|
-
|
|
16
|
-
# Test it
|
|
17
|
-
50c hints "api design patterns"
|
|
18
|
-
```
|
|
19
|
-
|
|
20
|
-
## MCP Integration
|
|
21
|
-
|
|
22
|
-
50c is an MCP (Model Context Protocol) server. After install, your AI IDE can use all tools directly.
|
|
23
|
-
|
|
24
|
-
**Supported IDEs:** Claude Desktop, Cursor, Windsurf, VS Code,
|
|
25
|
-
|
|
26
|
-
### Manual Configuration
|
|
27
|
-
|
|
28
|
-
Add to your IDE's MCP config
|
|
29
|
-
|
|
30
|
-
```json
|
|
31
|
-
{
|
|
32
|
-
"mcpServers": {
|
|
33
|
-
"50c": {
|
|
34
|
-
"command": "npx",
|
|
35
|
-
"args": ["-y", "50c@latest"],
|
|
36
|
-
"env": {
|
|
37
|
-
"FIFTYC_API_KEY": "YOUR_API_KEY"
|
|
38
|
-
}
|
|
39
|
-
}
|
|
40
|
-
}
|
|
41
|
-
}
|
|
42
|
-
```
|
|
43
|
-
|
|
44
|
-
##
|
|
45
|
-
|
|
46
|
-
|
|
47
|
-
|
|
48
|
-
|
|
49
|
-
|
|
50
|
-
|
|
51
|
-
|
|
52
|
-
|
|
53
|
-
|
|
54
|
-
|
|
55
|
-
|
|
56
|
-
|
|
57
|
-
|
|
58
|
-
|
|
59
|
-
|
|
60
|
-
|
|
61
|
-
|
|
62
|
-
|
|
63
|
-
|
|
64
|
-
|
|
65
|
-
|
|
66
|
-
|
|
67
|
-
|
|
68
|
-
|
|
69
|
-
|
|
70
|
-
|
|
71
|
-
|
|
72
|
-
|
|
|
73
|
-
|
|
74
|
-
| `
|
|
75
|
-
|
|
76
|
-
|
|
77
|
-
|
|
78
|
-
|
|
|
79
|
-
|
|
80
|
-
| `
|
|
81
|
-
| `
|
|
82
|
-
| `
|
|
83
|
-
| `
|
|
84
|
-
| `
|
|
85
|
-
|
|
86
|
-
|
|
87
|
-
|
|
88
|
-
|
|
89
|
-
|
|
90
|
-
|
|
91
|
-
|
|
|
92
|
-
|
|
93
|
-
| `
|
|
94
|
-
| `
|
|
95
|
-
| `
|
|
96
|
-
|
|
97
|
-
|
|
98
|
-
|
|
99
|
-
|
|
100
|
-
|
|
101
|
-
|
|
102
|
-
|
|
103
|
-
|
|
104
|
-
|
|
105
|
-
|
|
106
|
-
|
|
107
|
-
|
|
108
|
-
|
|
109
|
-
|
|
110
|
-
|
|
111
|
-
|
|
112
|
-
-
|
|
113
|
-
|
|
114
|
-
|
|
115
|
-
|
|
116
|
-
|
|
117
|
-
|
|
118
|
-
|
|
119
|
-
|
|
120
|
-
|
|
121
|
-
|
|
122
|
-
|
|
123
|
-
|
|
124
|
-
|
|
125
|
-
|
|
126
|
-
|
|
127
|
-
|
|
128
|
-
|
|
129
|
-
|
|
130
|
-
|
|
131
|
-
|
|
132
|
-
|
|
133
|
-
|
|
134
|
-
|
|
135
|
-
|
|
136
|
-
|
|
137
|
-
|
|
138
|
-
|
|
139
|
-
|
|
140
|
-
|
|
141
|
-
|
|
142
|
-
|
|
143
|
-
|
|
144
|
-
|
|
145
|
-
|
|
146
|
-
|
|
147
|
-
|
|
148
|
-
|
|
149
|
-
|
|
150
|
-
|
|
151
|
-
|
|
152
|
-
|
|
153
|
-
|
|
154
|
-
|
|
155
|
-
|
|
156
|
-
|
|
157
|
-
|
|
158
|
-
|
|
159
|
-
|
|
160
|
-
|
|
161
|
-
|
|
162
|
-
|
|
163
|
-
|
|
164
|
-
|
|
165
|
-
|
|
166
|
-
|
|
167
|
-
|
|
168
|
-
|
|
169
|
-
|
|
170
|
-
|
|
171
|
-
|
|
172
|
-
|
|
173
|
-
|
|
174
|
-
|
|
175
|
-
|
|
1
|
+
# 50c
|
|
2
|
+
|
|
3
|
+
AI developer tools via MCP. Pay-per-use from $0.01. No subscriptions.
|
|
4
|
+
|
|
5
|
+
**Requirements:** Node.js 18+
|
|
6
|
+
|
|
7
|
+
## Quick Start
|
|
8
|
+
|
|
9
|
+
```bash
|
|
10
|
+
# Install and configure for your IDE
|
|
11
|
+
npx 50c install
|
|
12
|
+
|
|
13
|
+
# Set API key (get one at https://50c.ai)
|
|
14
|
+
50c config key YOUR_API_KEY
|
|
15
|
+
|
|
16
|
+
# Test it
|
|
17
|
+
50c hints "api design patterns"
|
|
18
|
+
```
|
|
19
|
+
|
|
20
|
+
## MCP Integration
|
|
21
|
+
|
|
22
|
+
50c is an MCP (Model Context Protocol) server. After install, your AI IDE can use all tools directly.
|
|
23
|
+
|
|
24
|
+
**Supported IDEs:** Claude Desktop, Claude Code, Cursor, Windsurf, VS Code, Cline, Continue, Roo Code
|
|
25
|
+
|
|
26
|
+
### Manual Configuration
|
|
27
|
+
|
|
28
|
+
Add to your IDE's MCP config:
|
|
29
|
+
|
|
30
|
+
```json
|
|
31
|
+
{
|
|
32
|
+
"mcpServers": {
|
|
33
|
+
"50c": {
|
|
34
|
+
"command": "npx",
|
|
35
|
+
"args": ["-y", "50c@latest"],
|
|
36
|
+
"env": {
|
|
37
|
+
"FIFTYC_API_KEY": "YOUR_API_KEY"
|
|
38
|
+
}
|
|
39
|
+
}
|
|
40
|
+
}
|
|
41
|
+
}
|
|
42
|
+
```
|
|
43
|
+
|
|
44
|
+
## Security Audit Tool (NEW)
|
|
45
|
+
|
|
46
|
+
Built-in security scanner that checks your machine for backdoors, unauthorized access, and persistence mechanisms. Runs entirely locally. FREE.
|
|
47
|
+
|
|
48
|
+
```
|
|
49
|
+
backdoor_check
|
|
50
|
+
```
|
|
51
|
+
|
|
52
|
+
Cross-platform: Windows (17 checks), macOS (22 checks), Linux (13 checks).
|
|
53
|
+
|
|
54
|
+
**What it checks:**
|
|
55
|
+
- RDP/SSH login history with geo-IP tagging of foreign IPs
|
|
56
|
+
- Persistence mechanisms (Registry Run keys, LaunchAgents/Daemons, crontabs, systemd)
|
|
57
|
+
- User accounts and admin group membership
|
|
58
|
+
- Active network connections and listening ports
|
|
59
|
+
- Firewall rules and remote management (WinRM, VNC/ARD)
|
|
60
|
+
- Root certificates and proxy/MITM settings
|
|
61
|
+
- Suspicious IDE artifacts (Verdant, Trae, etc.)
|
|
62
|
+
- SSH authorized_keys anomalies
|
|
63
|
+
- Shell history for suspicious commands
|
|
64
|
+
- macOS: Gatekeeper, SIP, TCC privacy permissions, MDM profiles
|
|
65
|
+
|
|
66
|
+
Returns structured JSON with severity-rated alerts and a verdict (CLEAN / REVIEW / INVESTIGATE).
|
|
67
|
+
|
|
68
|
+
## Tools Overview
|
|
69
|
+
|
|
70
|
+
### Free Tools (No API Key Required)
|
|
71
|
+
|
|
72
|
+
| Tool | Description |
|
|
73
|
+
|------|-------------|
|
|
74
|
+
| `web_search` | Internet search with structured results |
|
|
75
|
+
| `page_fetch` | Fetch and extract URL content |
|
|
76
|
+
| `domain_check` | Check domain availability |
|
|
77
|
+
| `contrast_check` | WCAG color contrast checker |
|
|
78
|
+
| `backdoor_check` | Security audit (RDP/SSH, persistence, geo-IP) |
|
|
79
|
+
| `clarity` | Clarity Mode: tools self-activate based on input relevance |
|
|
80
|
+
| `skeptic_pov` | 5-lens adversarial verification |
|
|
81
|
+
| `solve` | Deep iterative solver with self-correction |
|
|
82
|
+
| `fm_index/find/search/lines/summary/list/context` | Fast file indexing and search |
|
|
83
|
+
| `roadmap_add/list/update/delete` | Dev roadmap tracking |
|
|
84
|
+
| `adoption_calc/diagnose/simulate` | Adoption probability modeling |
|
|
85
|
+
|
|
86
|
+
### Starter Tools ($0.01-$0.05)
|
|
87
|
+
|
|
88
|
+
| Tool | Cost | Description |
|
|
89
|
+
|------|------|-------------|
|
|
90
|
+
| `hints` | $0.05 | 5 brutal 2-word debugging hints |
|
|
91
|
+
| `roast` | $0.05 | Brutal code review with fixes |
|
|
92
|
+
| `name_it` | $0.03 | 5 product names + domain check |
|
|
93
|
+
| `beacon_compress` | $0.02 | Smart context compression |
|
|
94
|
+
| `beacon_extract` | $0.01 | Extract decisions/errors/entities |
|
|
95
|
+
| `fog_check` | $0.05 | Check conversation fog level |
|
|
96
|
+
|
|
97
|
+
### Pro Tools ($0.08-$0.20)
|
|
98
|
+
|
|
99
|
+
| Tool | Cost | Description |
|
|
100
|
+
|------|------|-------------|
|
|
101
|
+
| `genius` | $0.50 | Deep problem solving |
|
|
102
|
+
| `genius_plus` | $0.65 | Self-improving code generation |
|
|
103
|
+
| `compute` | $0.10 | Execute Python in sandbox |
|
|
104
|
+
| `justice` | $0.10 | Multi-framework ethical reasoning |
|
|
105
|
+
| `artist_mode` | $0.10 | Divergent creative construction |
|
|
106
|
+
| `self_modeling` | $0.08 | Epistemic self-analysis |
|
|
107
|
+
| `seo_audit` | $0.10 | Quick SEO audit |
|
|
108
|
+
| `prompt_fortress` | $0.20 | Prompt security stress-test |
|
|
109
|
+
|
|
110
|
+
### Titan Deep Solvers (Foundation Tier)
|
|
111
|
+
|
|
112
|
+
Purpose-built multi-phase solvers for hard mathematical and reasoning problems. Each Titan combines multiple 50c tools in orchestrated pipelines with ground-truth verification.
|
|
113
|
+
|
|
114
|
+
| Solver | Cost | Phases | Best For |
|
|
115
|
+
|--------|------|--------|----------|
|
|
116
|
+
| `titan_1` | $10 | 7: Diverge, Ground-truth, Explore, Attack, Resolve, Repair, Verify | Hardest problems. Full adversarial pipeline. |
|
|
117
|
+
| `titan_2` | $5 | 4: Ground-truth, Analysis, Synthesis, Scoring | Math needing compute-anchored verification. |
|
|
118
|
+
| `titan_3` | $10 | 6: Understand, Construct, Verify, Audit, Fix, Synthesize | Proof-heavy problems. |
|
|
119
|
+
| `titan_4` | $2 | 5: Blast, Verify, Fix, Synthesize, Claim Audit | Fast with auto-correction. |
|
|
120
|
+
|
|
121
|
+
**Titan Design Principles:**
|
|
122
|
+
- Compute-before-assert: numerical claims verified by Python before inclusion
|
|
123
|
+
- Ground-truth anchoring: Phase 1 runs brute-force computation to establish facts
|
|
124
|
+
- Claim audit: final phase extracts claims and compares against ground truth
|
|
125
|
+
- Self-correction: auto-corrects when claims don't match computation
|
|
126
|
+
|
|
127
|
+
**Benchmarks (External AGI-2035 Grading):**
|
|
128
|
+
- Titan-2 Q10: B+ - "first submission with appropriate intellectual honesty"
|
|
129
|
+
- Titan-3 Q11: A- at $10/358s - first correct proof across 5 models
|
|
130
|
+
|
|
131
|
+
### Clarity Mode
|
|
132
|
+
|
|
133
|
+
Intelligent tool auto-selection. Describe your problem and Clarity activates the right tools:
|
|
134
|
+
|
|
135
|
+
```
|
|
136
|
+
clarity({ input: "Is this proof valid?", mode: "auto", budget_cents: 200 })
|
|
137
|
+
```
|
|
138
|
+
|
|
139
|
+
### Skeptic POV
|
|
140
|
+
|
|
141
|
+
Free adversarial verification through 5 lenses: Factual, Logical, Computational, Constructive, STEM.
|
|
142
|
+
|
|
143
|
+
### Magnum Mathematical Verification Suite
|
|
144
|
+
|
|
145
|
+
| Tool | Cost | Description |
|
|
146
|
+
|------|------|-------------|
|
|
147
|
+
| `magnum_verify_proof` | $0.35 | Proof verification |
|
|
148
|
+
| `magnum_find_error` | $0.30 | Error detection in proofs |
|
|
149
|
+
| `magnum_number_theory` | $0.35 | Number theory specialist |
|
|
150
|
+
| `magnum_test_conjecture` | $0.35 | Adversarial conjecture testing |
|
|
151
|
+
| `magnum_find_theorem` | $0.30 | Find relevant theorems |
|
|
152
|
+
| `magnum_design_computation` | $0.30 | Design computational experiments |
|
|
153
|
+
| `magnum_dependency_dag` | $0.25 | Proof dependency graph |
|
|
154
|
+
| `magnum_symbolic_computational_hybrid` | $0.40 | Hybrid symbolic + computational |
|
|
155
|
+
| `mcp_conductor` | $0.50 | Meta-orchestrator for Magnum suite |
|
|
156
|
+
|
|
157
|
+
## 50c Team
|
|
158
|
+
|
|
159
|
+
Natural language orchestration - describe what you want, Team picks the right tools:
|
|
160
|
+
|
|
161
|
+
```
|
|
162
|
+
"roast my code and suggest fixes" -> roast + hints
|
|
163
|
+
"check if coolstartup.ai is available" -> domain_check
|
|
164
|
+
```
|
|
165
|
+
|
|
166
|
+
### SSH Server Configuration
|
|
167
|
+
|
|
168
|
+
Configure servers in `~/.50c/servers.json`:
|
|
169
|
+
|
|
170
|
+
```json
|
|
171
|
+
{
|
|
172
|
+
"prod": { "host": "10.0.0.1", "user": "deploy", "name": "Production" }
|
|
173
|
+
}
|
|
174
|
+
```
|
|
175
|
+
|
|
176
|
+
## Pre-Publish Verification
|
|
177
|
+
|
|
178
|
+
```bash
|
|
179
|
+
50c pre_publish npm # npm package
|
|
180
|
+
50c pre_publish arxiv # arXiv paper
|
|
181
|
+
50c pre_publish medical # Clinical paper
|
|
182
|
+
50c pre_publish github # GitHub release
|
|
183
|
+
```
|
|
184
|
+
|
|
185
|
+
## Pricing Tiers
|
|
186
|
+
|
|
187
|
+
| Tier | Price | Includes |
|
|
188
|
+
|------|-------|----------|
|
|
189
|
+
| Free | $0 | 40+ tools: search, file indexing, security audit, fog detection |
|
|
190
|
+
| Starter | $0.01-$0.05 | Beacon, hints, roast, naming |
|
|
191
|
+
| Pro | $0.08-$0.65 | Genius, compute, SEO, UX, ethics |
|
|
192
|
+
| Enterprise | $0.15-$10 | Titan solvers, Magnum math, CVI |
|
|
193
|
+
| Foundation | Custom | All tools, zero marginal cost |
|
|
194
|
+
|
|
195
|
+
Credits never expire. No subscriptions.
|
|
196
|
+
|
|
197
|
+
## Links
|
|
198
|
+
|
|
199
|
+
- Website: https://50c.ai
|
|
200
|
+
- Docs: https://docs.50c.ai
|
|
201
|
+
- Dashboard: https://50c.ai/dashboard.html
|