specflow-cc 1.6.0 → 1.6.2
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 +22 -0
- package/agents/codebase-scanner.md +4 -1
- package/package.json +1 -1
- package/templates/scan.md +2 -1
package/README.md
CHANGED
|
@@ -17,6 +17,10 @@ npx specflow-cc --global
|
|
|
17
17
|
|
|
18
18
|
<br>
|
|
19
19
|
|
|
20
|
+
https://github.com/user-attachments/assets/3f516907-8657-4ea4-bc0c-6319998a09db
|
|
21
|
+
|
|
22
|
+
<br>
|
|
23
|
+
|
|
20
24
|
*"The audit caught 4 issues before I wrote a single line of code."*
|
|
21
25
|
|
|
22
26
|
*"Fresh context review found a bug I would have shipped to production."*
|
|
@@ -48,6 +52,24 @@ The result: fewer surprises, cleaner code, and a paper trail of decisions.
|
|
|
48
52
|
|
|
49
53
|
---
|
|
50
54
|
|
|
55
|
+
## Why SpecFlow?
|
|
56
|
+
|
|
57
|
+
Most AI coding workflows suffer from **"Yes-Man Syndrome"** — you ask for a feature, the AI nods and writes code. 500 lines later, you discover hallucinated APIs, missing edge cases, or security holes.
|
|
58
|
+
|
|
59
|
+
SpecFlow fixes this with **Fresh Context Auditing**:
|
|
60
|
+
|
|
61
|
+
| Aspect | Standard AI Coding | SpecFlow |
|
|
62
|
+
|--------|-------------------|----------|
|
|
63
|
+
| **Context** | Single window (grows, degrades) | Fresh context per phase |
|
|
64
|
+
| **Quality Control** | Self-correction (biased) | Independent Auditor (unbiased) |
|
|
65
|
+
| **Verification** | "Looks good to me" | Verified against contract |
|
|
66
|
+
| **Execution** | Linear, single agent | Atomic waves, parallel agents |
|
|
67
|
+
| **Result** | Hidden bugs ship | Issues caught before code exists |
|
|
68
|
+
|
|
69
|
+
> The auditor has no memory of your conversation. It only sees the spec. If the spec says "use the flux capacitor API" — the auditor asks "what is that?"
|
|
70
|
+
|
|
71
|
+
---
|
|
72
|
+
|
|
51
73
|
## Who This Is For
|
|
52
74
|
|
|
53
75
|
Developers who want Claude Code to:
|
|
@@ -187,9 +187,12 @@ Based on this scan, consider creating specs for:
|
|
|
187
187
|
1. **[Spec title]** — [brief description]
|
|
188
188
|
- Priority: [High | Medium | Low]
|
|
189
189
|
- Complexity: [small | medium | large]
|
|
190
|
+
Run: `/sf:new "[title]"`
|
|
190
191
|
|
|
191
192
|
2. **[Spec title]** — [brief description]
|
|
192
|
-
|
|
193
|
+
- Priority: [High | Medium | Low]
|
|
194
|
+
- Complexity: [small | medium | large]
|
|
195
|
+
Run: `/sf:new "[title]"`
|
|
193
196
|
|
|
194
197
|
---
|
|
195
198
|
|
package/package.json
CHANGED
package/templates/scan.md
CHANGED
|
@@ -85,11 +85,12 @@ Based on this scan, consider creating specs for:
|
|
|
85
85
|
1. **[Spec title]** — [brief description]
|
|
86
86
|
- Priority: high | medium | low
|
|
87
87
|
- Complexity: small | medium | large
|
|
88
|
-
|
|
88
|
+
Run: `/sf:new "[title]"`
|
|
89
89
|
|
|
90
90
|
2. **[Spec title]** — [brief description]
|
|
91
91
|
- Priority: ...
|
|
92
92
|
- Complexity: ...
|
|
93
|
+
Run: `/sf:new "[title]"`
|
|
93
94
|
|
|
94
95
|
---
|
|
95
96
|
|