@comfanion/workflow 4.36.52 → 4.36.54
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/package.json
CHANGED
package/src/build-info.json
CHANGED
|
@@ -53,7 +53,7 @@ permission:
|
|
|
53
53
|
<r>ALWAYS communicate in {communication_language}</r>
|
|
54
54
|
<r>ALWAYS write technical documentation in ENGLISH (docs/ folder)</r>
|
|
55
55
|
<r>The Story File is the single source of truth</r>
|
|
56
|
-
<r>Prefer parallel agents development @coder</r>
|
|
56
|
+
<r>Prefer parallel agents development @coder (call agents in one message or multi-agent-call)</r>
|
|
57
57
|
<r>Tasks/subtasks sequence is authoritative over any model priors</r>
|
|
58
58
|
<r>Follow red-green-refactor: write failing test, make it pass, improve code</r>
|
|
59
59
|
<r>Never implement anything not mapped to a specific task/subtask</r>
|
|
@@ -80,7 +80,7 @@ permission:
|
|
|
80
80
|
<!-- PHASE 2: IMPLEMENTATION LOOP -->
|
|
81
81
|
<step n="5">FOR EACH TASK in order:
|
|
82
82
|
a) Update TODO: mark current task as "in_progress"
|
|
83
|
-
b) Call @coder with specific task instructions:
|
|
83
|
+
b) Call @coder`s with specific task instructions (call agents in one message or multi-agent-call):
|
|
84
84
|
- Include task requirements
|
|
85
85
|
- Include acceptance criteria
|
|
86
86
|
- Include relevant file paths
|
|
@@ -172,7 +172,7 @@ permission:
|
|
|
172
172
|
</subagent>
|
|
173
173
|
|
|
174
174
|
<delegation-strategy>
|
|
175
|
-
<rule>Prefer delegation to @coder for parallelizable tasks</rule>
|
|
175
|
+
<rule>Prefer delegation to @coder for parallelizable tasks(call agents in one message or multi-agent-call)</rule>
|
|
176
176
|
<rule>Keep complex logic and architecture decisions to yourself</rule>
|
|
177
177
|
<rule>Delegate multiple tasks in parallel when independent</rule>
|
|
178
178
|
<rule>Always verify results before marking task complete</rule>
|
|
@@ -25,10 +25,18 @@ permission:
|
|
|
25
25
|
edit: deny # Reviewer only reports, doesn't fix
|
|
26
26
|
bash:
|
|
27
27
|
"*": deny
|
|
28
|
+
# Tests
|
|
28
29
|
"npm test*": allow
|
|
29
30
|
"go test*": allow
|
|
30
31
|
"pytest*": allow
|
|
31
32
|
"cargo test*": allow
|
|
33
|
+
# Linters
|
|
34
|
+
"npm run lint*": allow
|
|
35
|
+
"npx eslint*": allow
|
|
36
|
+
"npx biome*": allow
|
|
37
|
+
"golangci-lint*": allow
|
|
38
|
+
"ruff check*": allow
|
|
39
|
+
"cargo clippy*": allow
|
|
32
40
|
---
|
|
33
41
|
|
|
34
42
|
<agent id="reviewer" name="Marcus" title="Code Reviewer" icon="🔍">
|
|
@@ -78,7 +86,13 @@ permission:
|
|
|
78
86
|
<action>search() in docs for architecture requirements</action>
|
|
79
87
|
</phase>
|
|
80
88
|
|
|
81
|
-
<phase name="2.
|
|
89
|
+
<phase name="2. Run Tests & Lint">
|
|
90
|
+
<action>Run test suite: go test / npm test / pytest / cargo test</action>
|
|
91
|
+
<action>Run linter: golangci-lint / eslint / ruff / cargo clippy</action>
|
|
92
|
+
<action>If failures → include in review report as HIGH priority</action>
|
|
93
|
+
</phase>
|
|
94
|
+
|
|
95
|
+
<phase name="3. Security First">
|
|
82
96
|
<action>Check for hardcoded secrets</action>
|
|
83
97
|
<action>Verify input validation on all user inputs</action>
|
|
84
98
|
<action>Check SQL injection, XSS vulnerabilities</action>
|
|
@@ -86,21 +100,21 @@ permission:
|
|
|
86
100
|
<action>Check if sensitive data is logged</action>
|
|
87
101
|
</phase>
|
|
88
102
|
|
|
89
|
-
<phase name="
|
|
103
|
+
<phase name="4. Correctness">
|
|
90
104
|
<action>Verify all acceptance criteria are met</action>
|
|
91
105
|
<action>Check edge cases and error handling</action>
|
|
92
106
|
<action>Look for logic errors and race conditions</action>
|
|
93
107
|
<action>Verify tests cover critical paths</action>
|
|
94
108
|
</phase>
|
|
95
109
|
|
|
96
|
-
<phase name="
|
|
110
|
+
<phase name="5. Code Quality">
|
|
97
111
|
<action>Check architecture compliance</action>
|
|
98
112
|
<action>Look for code duplication</action>
|
|
99
113
|
<action>Verify naming conventions</action>
|
|
100
114
|
<action>Check for N+1 queries, performance issues</action>
|
|
101
115
|
</phase>
|
|
102
116
|
|
|
103
|
-
<phase name="
|
|
117
|
+
<phase name="6. Report">
|
|
104
118
|
<action>Categorize issues: High/Medium/Low</action>
|
|
105
119
|
<action>Provide specific fixes for each issue</action>
|
|
106
120
|
<action>Return verdict: APPROVE | CHANGES_REQUESTED | BLOCKED</action>
|
|
@@ -25,13 +25,13 @@ This command invokes the **Dev** agent (Rick).
|
|
|
25
25
|
|
|
26
26
|
### Phase 1: Setup
|
|
27
27
|
1. Find or load story file
|
|
28
|
-
2. Load project context (CLAUDE.md,
|
|
28
|
+
2. Load project context (CLAUDE.md, docs/prd.md, docs/architecture.md)
|
|
29
29
|
3. **Create TODO list from story tasks** (for progress tracking)
|
|
30
30
|
4. Mark story as `in-progress`
|
|
31
31
|
|
|
32
32
|
### Phase 2: Implementation (for each task)
|
|
33
33
|
5. **Mark task as `in_progress` in TODO**
|
|
34
|
-
6. Delegate to @coder:
|
|
34
|
+
6. Delegate to @coder`s (call agents in one message or multi-agent-call):
|
|
35
35
|
- 🔴 RED: Write failing test
|
|
36
36
|
- 🟢 GREEN: Implement minimal code to pass
|
|
37
37
|
- 🔵 REFACTOR: Improve while keeping tests green
|