@brokkai/bug-bot 0.3.2 → 0.3.4

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.
Files changed (2) hide show
  1. package/README.md +33 -5
  2. package/package.json +5 -5
package/README.md CHANGED
@@ -50,6 +50,7 @@ make build
50
50
  ./bin/bbb /path/to/your-repo --max-issues 2 --label bug
51
51
  ./bin/bbb /path/to/your-repo --model YOUR_MODEL_ID --effort low
52
52
  ./bin/bbb status /path/to/your-repo
53
+ ./bin/bbb report /path/to/your-repo --status dry_run > findings.md
53
54
  ./bin/bbb version
54
55
  ./bin/bbb retry /path/to/your-repo --once
55
56
  ```
@@ -108,7 +109,7 @@ details on exit.
108
109
 
109
110
  Piped input, redirected stderr, and `TERM=dumb` use scrolling output automatically.
110
111
  `--plain` and `--json` disable the dashboard and are mutually exclusive.
111
- `NO_COLOR` disables dashboard colors. `status`, `version`, and help keep their
112
+ `NO_COLOR` disables dashboard colors. `status`, `report`, `version`, and help keep their
112
113
  existing output and never open the dashboard.
113
114
 
114
115
  ## How it works
@@ -144,9 +145,15 @@ and another scan **30 minutes after completion**, even if the commit is unchange
144
145
  Recent summaries guide exploration; this is not a claim of exhaustive coverage.
145
146
  `once` runs or resumes one scan and exits. Failed scans retain their candidates,
146
147
  workspace, and diagnostics; retries wait at least 15 minutes and run on the next
147
- poll, with three attempts before requiring `retry`. Agent setup errors stop the
148
- daemon without consuming an attempt. An advanced branch invalidates pending
149
- findings so the next eligible attempt scans the new commit.
148
+ poll, with three attempts before requiring `retry`. Agent startup failures are
149
+ retried in place three times within about a minute, since nothing has been
150
+ prompted yet; persistent setup errors then stop the daemon without consuming
151
+ an attempt. After reconciling any uncertain issue
152
+ publications, each poll or one-shot invocation checks for an advanced branch
153
+ before enforcing the saved revision’s attempt budget or retry delay. A new
154
+ revision archives pending findings as stale, retains their evidence and workspace,
155
+ and starts a separate scan with a fresh budget. An unchanged exhausted revision
156
+ remains blocked until `retry`.
150
157
 
151
158
  ## Duplicate handling and interrupted requests
152
159
 
@@ -201,7 +208,9 @@ TLS transport to be added later without changing worker semantics.
201
208
  ## Optional configuration
202
209
 
203
210
  `bbb --config bug-bot.json` loads a strict JSON object. No file is loaded or
204
- generated implicitly. Paths are resolved relative to the configuration file.
211
+ generated implicitly. Paths, including local mirror remotes such as
212
+ `./mirror.git` or `./published:mirror.git`, are resolved relative to the
213
+ configuration file.
205
214
  See [bug-bot.example.json](bug-bot.example.json).
206
215
 
207
216
  ```json
@@ -239,6 +248,25 @@ remote and branch. JSON state is replaced atomically with fsync; private session
239
248
  transcripts live under the state directory. `status` prints saved JSON without
240
249
  starting an agent. `--json` selects structured progress logs.
241
250
 
251
+ Export all retained findings as Markdown for later review or sharing:
252
+
253
+ ```sh
254
+ bbb report --config bug-bot.json > findings.md
255
+ bbb report --config bug-bot.json --status dry_run > dry-run.md
256
+ ```
257
+
258
+ `report` reads saved state without starting an agent or scan, or changing state.
259
+ A repository argument and `--branch` use the same selection as other commands;
260
+ `--config` avoids remote discovery and requires neither GitHub CLI nor an agent.
261
+ The report includes every completed and active finding (without the dashboard's
262
+ 200-entry limit), evidence, review, saved status, and any issue URL. Filter with
263
+ `--status pending|posting|submitted|duplicate|uncertain|invalid|dry_run|stale`
264
+ (select one value); omitting it includes all outcomes. Empty selections report
265
+ no saved findings. Active findings show the scan's known commit; completed
266
+ findings explicitly show commit unavailable because it is not retained. Internal
267
+ checkout/state paths and publication request markers are omitted from report
268
+ metadata; saved finding text is preserved.
269
+
242
270
  Scan worktrees and reproduction files are retained for inspection. Manage their
243
271
  retention along with transcripts externally. Agent instructions prohibit fixes,
244
272
  commits, pushes, and direct GitHub writes; tracked source changes or a changed
package/package.json CHANGED
@@ -1,5 +1,5 @@
1
1
  {
2
- "version": "0.3.2",
2
+ "version": "0.3.4",
3
3
  "license": "Apache-2.0",
4
4
  "repository": {
5
5
  "type": "git",
@@ -25,9 +25,9 @@
25
25
  "arm64"
26
26
  ],
27
27
  "optionalDependencies": {
28
- "@brokkai/bug-bot-linux-x64": "0.3.2",
29
- "@brokkai/bug-bot-linux-arm64": "0.3.2",
30
- "@brokkai/bug-bot-darwin-x64": "0.3.2",
31
- "@brokkai/bug-bot-darwin-arm64": "0.3.2"
28
+ "@brokkai/bug-bot-linux-x64": "0.3.4",
29
+ "@brokkai/bug-bot-linux-arm64": "0.3.4",
30
+ "@brokkai/bug-bot-darwin-x64": "0.3.4",
31
+ "@brokkai/bug-bot-darwin-arm64": "0.3.4"
32
32
  }
33
33
  }