clud-bug 0.5.2 โ 0.5.4
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 +20 -0
- package/package.json +1 -1
- package/templates/workflow-py.yml.tmpl +34 -0
- package/templates/workflow-ts.yml.tmpl +34 -0
- package/templates/workflow.yml.tmpl +34 -0
package/README.md
CHANGED
|
@@ -182,6 +182,26 @@ After install:
|
|
|
182
182
|
3. Within ~2 min, Clud Bug should post a comment flagging it.
|
|
183
183
|
4. If no comment: check the **Actions** tab logs. Look for `gh pr comment` invocations and any "Resource not accessible by integration" errors (usually a permissions issue or a fork PR).
|
|
184
184
|
|
|
185
|
+
### Reading a review
|
|
186
|
+
|
|
187
|
+
Every Clud Bug review opens with a status line that tells you exactly what changed since the previous pass โ particularly useful on re-review after you push a fix:
|
|
188
|
+
|
|
189
|
+
```
|
|
190
|
+
## ๐ Clud Bug review
|
|
191
|
+
|
|
192
|
+
**This round:** 0 critical ยท 1 minor ยท 3 resolved from prior ยท 0 still open
|
|
193
|
+
|
|
194
|
+
### Findings
|
|
195
|
+
โฆ
|
|
196
|
+
```
|
|
197
|
+
|
|
198
|
+
- **critical** โ new critical findings in this review (these are what strict mode gates on)
|
|
199
|
+
- **minor** โ non-critical findings (suggestions / nits)
|
|
200
|
+
- **resolved from prior** โ prior unresolved threads the bot just cleared because it verified your fix in the diff
|
|
201
|
+
- **still open** โ prior threads whose issue is still standing
|
|
202
|
+
|
|
203
|
+
Same format every time; zero values are always present so the line is easy to scan and parse.
|
|
204
|
+
|
|
185
205
|
## Manual install (advanced)
|
|
186
206
|
|
|
187
207
|
If you don't want to use the CLI, you can install a generic workflow by hand:
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "clud-bug",
|
|
3
|
-
"version": "0.5.
|
|
3
|
+
"version": "0.5.4",
|
|
4
4
|
"description": "Claude PR review with project-aware skills. CLI installs a working GitHub Actions workflow and curates skills from skills.sh.",
|
|
5
5
|
"homepage": "https://cludbug.dev",
|
|
6
6
|
"bugs": "https://github.com/thrillmot/clud-bug/issues",
|
|
@@ -113,6 +113,40 @@ jobs:
|
|
|
113
113
|
|
|
114
114
|
## ๐ Clud Bug review
|
|
115
115
|
|
|
116
|
+
Immediately after the H2 header โ on the next non-empty
|
|
117
|
+
line โ emit a status block in this exact format:
|
|
118
|
+
|
|
119
|
+
**This round:** N critical ยท N minor ยท N resolved from prior ยท N still open
|
|
120
|
+
|
|
121
|
+
This applies to BOTH the bare "## ๐ Clud Bug review" header
|
|
122
|
+
and the strict-mode variants ("โ critical findings" /
|
|
123
|
+
"โ clean"). The status line goes on the next non-empty line
|
|
124
|
+
regardless of which header you used. Do not omit the H2
|
|
125
|
+
header variant in strict mode just to fit the status line โ
|
|
126
|
+
the strict-mode gate reads the H2 line and would break.
|
|
127
|
+
|
|
128
|
+
The four counters (always include all four, even when 0 โ
|
|
129
|
+
fixed format is grep-able and lets agents reading the
|
|
130
|
+
comment parse it deterministically):
|
|
131
|
+
โข critical โ count of NEW critical findings
|
|
132
|
+
in this review (the ones strict
|
|
133
|
+
mode gates on)
|
|
134
|
+
โข minor โ count of non-critical findings
|
|
135
|
+
(suggestions / nits / observations)
|
|
136
|
+
โข resolved from prior โ count of prior unresolved threads
|
|
137
|
+
YOU (claude[bot]) just resolved on
|
|
138
|
+
this pass via resolveReviewThread
|
|
139
|
+
(the loop-closing signal โ this
|
|
140
|
+
tells the author the bot read
|
|
141
|
+
their fixes)
|
|
142
|
+
โข still open โ count of prior unresolved threads
|
|
143
|
+
whose issue still stands AFTER
|
|
144
|
+
this pass
|
|
145
|
+
|
|
146
|
+
On a first-time review, "resolved from prior" and "still
|
|
147
|
+
open" are both 0. On follow-up reviews after a fix-push,
|
|
148
|
+
"resolved from prior" should typically be positive.
|
|
149
|
+
|
|
116
150
|
Post it via:
|
|
117
151
|
gh pr comment "$PR_NUMBER" --body "<your review>"
|
|
118
152
|
|
|
@@ -114,6 +114,40 @@ jobs:
|
|
|
114
114
|
|
|
115
115
|
## ๐ Clud Bug review
|
|
116
116
|
|
|
117
|
+
Immediately after the H2 header โ on the next non-empty
|
|
118
|
+
line โ emit a status block in this exact format:
|
|
119
|
+
|
|
120
|
+
**This round:** N critical ยท N minor ยท N resolved from prior ยท N still open
|
|
121
|
+
|
|
122
|
+
This applies to BOTH the bare "## ๐ Clud Bug review" header
|
|
123
|
+
and the strict-mode variants ("โ critical findings" /
|
|
124
|
+
"โ clean"). The status line goes on the next non-empty line
|
|
125
|
+
regardless of which header you used. Do not omit the H2
|
|
126
|
+
header variant in strict mode just to fit the status line โ
|
|
127
|
+
the strict-mode gate reads the H2 line and would break.
|
|
128
|
+
|
|
129
|
+
The four counters (always include all four, even when 0 โ
|
|
130
|
+
fixed format is grep-able and lets agents reading the
|
|
131
|
+
comment parse it deterministically):
|
|
132
|
+
โข critical โ count of NEW critical findings
|
|
133
|
+
in this review (the ones strict
|
|
134
|
+
mode gates on)
|
|
135
|
+
โข minor โ count of non-critical findings
|
|
136
|
+
(suggestions / nits / observations)
|
|
137
|
+
โข resolved from prior โ count of prior unresolved threads
|
|
138
|
+
YOU (claude[bot]) just resolved on
|
|
139
|
+
this pass via resolveReviewThread
|
|
140
|
+
(the loop-closing signal โ this
|
|
141
|
+
tells the author the bot read
|
|
142
|
+
their fixes)
|
|
143
|
+
โข still open โ count of prior unresolved threads
|
|
144
|
+
whose issue still stands AFTER
|
|
145
|
+
this pass
|
|
146
|
+
|
|
147
|
+
On a first-time review, "resolved from prior" and "still
|
|
148
|
+
open" are both 0. On follow-up reviews after a fix-push,
|
|
149
|
+
"resolved from prior" should typically be positive.
|
|
150
|
+
|
|
117
151
|
Post it via:
|
|
118
152
|
gh pr comment "$PR_NUMBER" --body "<your review>"
|
|
119
153
|
|
|
@@ -134,6 +134,40 @@ jobs:
|
|
|
134
134
|
|
|
135
135
|
## ๐ Clud Bug review
|
|
136
136
|
|
|
137
|
+
Immediately after the H2 header โ on the next non-empty
|
|
138
|
+
line โ emit a status block in this exact format:
|
|
139
|
+
|
|
140
|
+
**This round:** N critical ยท N minor ยท N resolved from prior ยท N still open
|
|
141
|
+
|
|
142
|
+
This applies to BOTH the bare "## ๐ Clud Bug review" header
|
|
143
|
+
and the strict-mode variants ("โ critical findings" /
|
|
144
|
+
"โ clean"). The status line goes on the next non-empty line
|
|
145
|
+
regardless of which header you used. Do not omit the H2
|
|
146
|
+
header variant in strict mode just to fit the status line โ
|
|
147
|
+
the strict-mode gate reads the H2 line and would break.
|
|
148
|
+
|
|
149
|
+
The four counters (always include all four, even when 0 โ
|
|
150
|
+
fixed format is grep-able and lets agents reading the
|
|
151
|
+
comment parse it deterministically):
|
|
152
|
+
โข critical โ count of NEW critical findings
|
|
153
|
+
in this review (the ones strict
|
|
154
|
+
mode gates on)
|
|
155
|
+
โข minor โ count of non-critical findings
|
|
156
|
+
(suggestions / nits / observations)
|
|
157
|
+
โข resolved from prior โ count of prior unresolved threads
|
|
158
|
+
YOU (claude[bot]) just resolved on
|
|
159
|
+
this pass via resolveReviewThread
|
|
160
|
+
(the loop-closing signal โ this
|
|
161
|
+
tells the author the bot read
|
|
162
|
+
their fixes)
|
|
163
|
+
โข still open โ count of prior unresolved threads
|
|
164
|
+
whose issue still stands AFTER
|
|
165
|
+
this pass
|
|
166
|
+
|
|
167
|
+
On a first-time review, "resolved from prior" and "still
|
|
168
|
+
open" are both 0. On follow-up reviews after a fix-push,
|
|
169
|
+
"resolved from prior" should typically be positive.
|
|
170
|
+
|
|
137
171
|
Post it via:
|
|
138
172
|
gh pr comment "$PR_NUMBER" --body "<your review>"
|
|
139
173
|
|