specflow-cc 1.5.1 → 1.5.3
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/CHANGELOG.md +21 -0
- package/agents/impl-reviewer.md +38 -4
- package/agents/spec-auditor.md +21 -1
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -5,6 +5,27 @@ All notable changes to SpecFlow will be documented in this file.
|
|
|
5
5
|
The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
|
|
6
6
|
and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
|
|
7
7
|
|
|
8
|
+
## [1.5.3] - 2026-01-22
|
|
9
|
+
|
|
10
|
+
### Changed
|
|
11
|
+
|
|
12
|
+
- `spec-auditor` now checks 8 quality dimensions (added: architecture fit, non-duplication, cognitive load)
|
|
13
|
+
- `impl-reviewer` now checks 8 quality dimensions (added: architecture, non-duplication, cognitive load)
|
|
14
|
+
- Enhanced integration checks for natural codebase fit
|
|
15
|
+
- Reviewers verify code doesn't reinvent existing solutions
|
|
16
|
+
- Both agents evaluate maintainability and developer experience
|
|
17
|
+
|
|
18
|
+
---
|
|
19
|
+
|
|
20
|
+
## [1.5.2] - 2026-01-22
|
|
21
|
+
|
|
22
|
+
### Fixed
|
|
23
|
+
|
|
24
|
+
- `/sf:review` now offers `/sf:fix` option when APPROVED with minor issues
|
|
25
|
+
- impl-reviewer agent output distinguishes "no issues" vs "with minor issues"
|
|
26
|
+
|
|
27
|
+
---
|
|
28
|
+
|
|
8
29
|
## [1.5.1] - 2026-01-22
|
|
9
30
|
|
|
10
31
|
### Added
|
package/agents/impl-reviewer.md
CHANGED
|
@@ -49,9 +49,12 @@ You audit with NO knowledge of HOW the code was written. This ensures:
|
|
|
49
49
|
|
|
50
50
|
1. **Compliance:** Does it meet the specification?
|
|
51
51
|
2. **Quality:** Is the code clean and maintainable?
|
|
52
|
-
3. **Integration:** Does it fit
|
|
52
|
+
3. **Integration:** Does it fit naturally with existing codebase?
|
|
53
53
|
4. **Security:** Are there vulnerabilities?
|
|
54
54
|
5. **Completeness:** Are all files/deletions handled?
|
|
55
|
+
6. **Architecture:** Does it follow established patterns?
|
|
56
|
+
7. **Non-duplication:** Does it reuse existing solutions where appropriate?
|
|
57
|
+
8. **Cognitive load:** Is it easy for other developers to understand?
|
|
55
58
|
|
|
56
59
|
</philosophy>
|
|
57
60
|
|
|
@@ -121,8 +124,29 @@ For each created/modified file:
|
|
|
121
124
|
### 4.4 Integration Check
|
|
122
125
|
|
|
123
126
|
- [ ] Matches existing code style
|
|
124
|
-
- [ ] Uses existing utilities (no reinventing)
|
|
125
127
|
- [ ] Proper imports/exports
|
|
128
|
+
- [ ] Fits naturally with surrounding code
|
|
129
|
+
|
|
130
|
+
### 4.5 Architecture Check
|
|
131
|
+
|
|
132
|
+
- [ ] Follows established patterns from PROJECT.md
|
|
133
|
+
- [ ] No conflicting architectural decisions
|
|
134
|
+
- [ ] Proper separation of concerns
|
|
135
|
+
- [ ] Dependencies flow in correct direction
|
|
136
|
+
|
|
137
|
+
### 4.6 Non-Duplication Check
|
|
138
|
+
|
|
139
|
+
- [ ] Uses existing utilities/helpers (no reinventing)
|
|
140
|
+
- [ ] No copy-paste of existing code that could be reused
|
|
141
|
+
- [ ] Leverages existing abstractions where appropriate
|
|
142
|
+
|
|
143
|
+
### 4.7 Cognitive Load Check
|
|
144
|
+
|
|
145
|
+
- [ ] Code is as simple as possible for the task
|
|
146
|
+
- [ ] Naming is clear and consistent with codebase
|
|
147
|
+
- [ ] No unnecessary abstractions or indirection
|
|
148
|
+
- [ ] Logic flow is easy to follow
|
|
149
|
+
- [ ] Future maintainers can understand without extensive context
|
|
126
150
|
|
|
127
151
|
## Step 5: Categorize Findings
|
|
128
152
|
|
|
@@ -246,9 +270,16 @@ Return review result:
|
|
|
246
270
|
|
|
247
271
|
## Next Step
|
|
248
272
|
|
|
249
|
-
{If APPROVED
|
|
273
|
+
{If APPROVED with NO minor issues:}
|
|
274
|
+
`/sf:done` — finalize and archive
|
|
250
275
|
|
|
251
|
-
{If
|
|
276
|
+
{If APPROVED WITH minor issues:}
|
|
277
|
+
Choose one:
|
|
278
|
+
• `/sf:done` — finalize as-is (minor issues are optional)
|
|
279
|
+
• `/sf:fix` — address minor issues first
|
|
280
|
+
|
|
281
|
+
{If CHANGES_REQUESTED:}
|
|
282
|
+
`/sf:fix` — address issues
|
|
252
283
|
|
|
253
284
|
Options:
|
|
254
285
|
- `/sf:fix all` — apply all fixes
|
|
@@ -265,6 +296,9 @@ Options:
|
|
|
265
296
|
- [ ] No lingering references to deleted code
|
|
266
297
|
- [ ] Code quality evaluated
|
|
267
298
|
- [ ] Security checked
|
|
299
|
+
- [ ] Architecture alignment verified
|
|
300
|
+
- [ ] No unnecessary duplication
|
|
301
|
+
- [ ] Cognitive load acceptable
|
|
268
302
|
- [ ] Findings categorized
|
|
269
303
|
- [ ] Review recorded in spec
|
|
270
304
|
- [ ] STATE.md updated
|
package/agents/spec-auditor.md
CHANGED
|
@@ -45,6 +45,9 @@ You are intentionally given NO context about how the spec was created. This ensu
|
|
|
45
45
|
3. **Testability:** Can each criterion be verified?
|
|
46
46
|
4. **Scope:** Is the boundary clear?
|
|
47
47
|
5. **Feasibility:** Is this achievable as specified?
|
|
48
|
+
6. **Architecture fit:** Does approach align with existing codebase patterns?
|
|
49
|
+
7. **Non-duplication:** Does this avoid reinventing existing solutions?
|
|
50
|
+
8. **Cognitive load:** Will this be easy for developers to understand and maintain?
|
|
48
51
|
|
|
49
52
|
</philosophy>
|
|
50
53
|
|
|
@@ -94,6 +97,23 @@ Evaluate each dimension:
|
|
|
94
97
|
- [ ] Assumptions are reasonable
|
|
95
98
|
- [ ] No impossible requirements
|
|
96
99
|
|
|
100
|
+
### Architecture Fit Check
|
|
101
|
+
- [ ] Approach aligns with existing codebase patterns
|
|
102
|
+
- [ ] Uses established conventions from PROJECT.md
|
|
103
|
+
- [ ] Integrates naturally with existing modules
|
|
104
|
+
- [ ] Doesn't introduce conflicting patterns
|
|
105
|
+
|
|
106
|
+
### Non-Duplication Check
|
|
107
|
+
- [ ] Doesn't duplicate existing functionality in codebase
|
|
108
|
+
- [ ] Reuses existing utilities/helpers where appropriate
|
|
109
|
+
- [ ] No "reinventing the wheel" when solution exists
|
|
110
|
+
|
|
111
|
+
### Cognitive Load Check
|
|
112
|
+
- [ ] Solution is as simple as possible for the task
|
|
113
|
+
- [ ] Naming is clear and consistent with codebase
|
|
114
|
+
- [ ] No unnecessary abstractions or indirection
|
|
115
|
+
- [ ] Future maintainers can understand the approach
|
|
116
|
+
|
|
97
117
|
## Step 4: Categorize Issues
|
|
98
118
|
|
|
99
119
|
Separate findings into:
|
|
@@ -188,7 +208,7 @@ Return formatted audit result:
|
|
|
188
208
|
<success_criteria>
|
|
189
209
|
- [ ] Specification fully read
|
|
190
210
|
- [ ] PROJECT.md context loaded
|
|
191
|
-
- [ ] All
|
|
211
|
+
- [ ] All 8 dimensions evaluated (clarity, completeness, testability, scope, feasibility, architecture, duplication, cognitive load)
|
|
192
212
|
- [ ] Issues categorized (critical vs recommendations)
|
|
193
213
|
- [ ] Audit recorded in spec's Audit History
|
|
194
214
|
- [ ] STATE.md updated
|