devrites 6.9.1 → 6.9.3
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 +12 -0
- package/README.md +1 -1
- package/pack/.claude/skills/devrites-lib/reference/parallel-dispatch.md +4 -1
- package/pack/.claude/skills/rite-autocomplete/SKILL.md +2 -1
- package/pack/generated/claude/skills/devrites-lib/reference/parallel-dispatch.md +4 -1
- package/pack/generated/claude/skills/rite-autocomplete/SKILL.md +2 -1
- package/pack/generated/codex/skills/devrites-lib/reference/parallel-dispatch.md +4 -1
- package/pack/generated/codex/skills/rite-autocomplete/SKILL.md +2 -1
- package/pack/generated/devin/skills/devrites-lib/reference/parallel-dispatch.md +4 -1
- package/pack/generated/devin/skills/rite-autocomplete/SKILL.md +2 -1
- package/pack/generated/omp/skills/devrites-lib/reference/parallel-dispatch.md +4 -1
- package/pack/generated/omp/skills/rite-autocomplete/SKILL.md +2 -1
- package/pack/generated/pi/skills/devrites-lib/reference/parallel-dispatch.md +4 -1
- package/pack/generated/pi/skills/rite-autocomplete/SKILL.md +2 -1
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,18 @@
|
|
|
2
2
|
|
|
3
3
|
All notable changes to DevRites are documented here. The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/) and DevRites adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html). Releases are generated automatically by [semantic-release](https://semantic-release.gitbook.io/) from Conventional Commits on `main`.
|
|
4
4
|
|
|
5
|
+
## [6.9.3](https://github.com/ViktorsBaikers/DevRites/compare/v6.9.2...v6.9.3) (2026-10-18)
|
|
6
|
+
|
|
7
|
+
### Fixed
|
|
8
|
+
|
|
9
|
+
* **devrites:** dispatch named roles only through the host agent tool ([#134](https://github.com/ViktorsBaikers/DevRites/issues/134)) ([60ca9d5](https://github.com/ViktorsBaikers/DevRites/commit/60ca9d59ed47758e403aa92ed033885b257270bd))
|
|
10
|
+
|
|
11
|
+
## [6.9.2](https://github.com/ViktorsBaikers/DevRites/compare/v6.9.1...v6.9.2) (2026-10-18)
|
|
12
|
+
|
|
13
|
+
### Fixed
|
|
14
|
+
|
|
15
|
+
* **devrites:** print the autocomplete card only at a final state ([#133](https://github.com/ViktorsBaikers/DevRites/issues/133)) ([fd1c5cf](https://github.com/ViktorsBaikers/DevRites/commit/fd1c5cfd89877de82d7ca2433704f578e40d5251))
|
|
16
|
+
|
|
5
17
|
## [6.9.1](https://github.com/ViktorsBaikers/DevRites/compare/v6.9.0...v6.9.1) (2026-10-18)
|
|
6
18
|
|
|
7
19
|
### Fixed
|
package/README.md
CHANGED
|
@@ -31,7 +31,7 @@ history can't lose it. After you clear the conversation, switch hosts, or hand
|
|
|
31
31
|
the feature to another engineer, the next session reads the same plan,
|
|
32
32
|
decisions, and evidence before it continues.
|
|
33
33
|
|
|
34
|
-
**Status:** [`v6.9.
|
|
34
|
+
**Status:** [`v6.9.3`](https://github.com/ViktorsBaikers/DevRites/releases/tag/v6.9.3): see [`CHANGELOG.md`](CHANGELOG.md) for release notes.
|
|
35
35
|
|
|
36
36
|
This is the latest published release; `main` may contain unreleased work.
|
|
37
37
|
|
|
@@ -51,7 +51,10 @@ If the candidate diff has no matching surfaces, record
|
|
|
51
51
|
a reviewer, auditor, or test-analyst wave adds `--packet <file>`, the exact packet
|
|
52
52
|
those agents receive, and `open` refuses one that seeds a gate result, outcome,
|
|
53
53
|
count, or "already passed" claim; at Review and Seal it also needs the packet to
|
|
54
|
-
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking
|
|
54
|
+
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking through the host's own agent
|
|
55
|
+
tool (never a shell-launched host CLI such as `omp -p` or `claude -p`: that is a
|
|
56
|
+
dispatch bridge, its PID handle dies with the session, and its return is a `gap`),
|
|
57
|
+
record each returned host handle
|
|
55
58
|
(`dispatch ... start --wave <w> --role <r> --handle <id>`), then `seal` —
|
|
56
59
|
seal fails until every declared role has a distinct handle, which makes a
|
|
57
60
|
serial launch-then-wait pattern impossible to record. Record each completion
|
|
@@ -149,7 +149,8 @@ keeps chaining pending slices. One Autocomplete invocation owns all internal
|
|
|
149
149
|
backtracking. Context pressure, compaction, or a nested completion footer does
|
|
150
150
|
not create a stop; persist and resume the cursor.
|
|
151
151
|
|
|
152
|
-
|
|
152
|
+
Reply only at a final state, never as mid-run progress: a text-only turn ends
|
|
153
|
+
the run, so after a compaction or phase change keep calling tools. Use one terse line:
|
|
153
154
|
|
|
154
155
|
```text
|
|
155
156
|
Autocomplete: <slug>
|
|
@@ -51,7 +51,10 @@ If the candidate diff has no matching surfaces, record
|
|
|
51
51
|
a reviewer, auditor, or test-analyst wave adds `--packet <file>`, the exact packet
|
|
52
52
|
those agents receive, and `open` refuses one that seeds a gate result, outcome,
|
|
53
53
|
count, or "already passed" claim; at Review and Seal it also needs the packet to
|
|
54
|
-
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking
|
|
54
|
+
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking through the host's own agent
|
|
55
|
+
tool (never a shell-launched host CLI such as `omp -p` or `claude -p`: that is a
|
|
56
|
+
dispatch bridge, its PID handle dies with the session, and its return is a `gap`),
|
|
57
|
+
record each returned host handle
|
|
55
58
|
(`dispatch ... start --wave <w> --role <r> --handle <id>`), then `seal` —
|
|
56
59
|
seal fails until every declared role has a distinct handle, which makes a
|
|
57
60
|
serial launch-then-wait pattern impossible to record. Record each completion
|
|
@@ -151,7 +151,8 @@ keeps chaining pending slices. One Autocomplete invocation owns all internal
|
|
|
151
151
|
backtracking. Context pressure, compaction, or a nested completion footer does
|
|
152
152
|
not create a stop; persist and resume the cursor.
|
|
153
153
|
|
|
154
|
-
|
|
154
|
+
Reply only at a final state, never as mid-run progress: a text-only turn ends
|
|
155
|
+
the run, so after a compaction or phase change keep calling tools. Use one terse line:
|
|
155
156
|
|
|
156
157
|
```text
|
|
157
158
|
Autocomplete: <slug>
|
|
@@ -51,7 +51,10 @@ If the candidate diff has no matching surfaces, record
|
|
|
51
51
|
a reviewer, auditor, or test-analyst wave adds `--packet <file>`, the exact packet
|
|
52
52
|
those agents receive, and `open` refuses one that seeds a gate result, outcome,
|
|
53
53
|
count, or "already passed" claim; at Review and Seal it also needs the packet to
|
|
54
|
-
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking
|
|
54
|
+
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking through the host's own agent
|
|
55
|
+
tool (never a shell-launched host CLI such as `omp -p` or `claude -p`: that is a
|
|
56
|
+
dispatch bridge, its PID handle dies with the session, and its return is a `gap`),
|
|
57
|
+
record each returned host handle
|
|
55
58
|
(`dispatch ... start --wave <w> --role <r> --handle <id>`), then `seal` —
|
|
56
59
|
seal fails until every declared role has a distinct handle, which makes a
|
|
57
60
|
serial launch-then-wait pattern impossible to record. Record each completion
|
|
@@ -151,7 +151,8 @@ keeps chaining pending slices. One Autocomplete invocation owns all internal
|
|
|
151
151
|
backtracking. Context pressure, compaction, or a nested completion footer does
|
|
152
152
|
not create a stop; persist and resume the cursor.
|
|
153
153
|
|
|
154
|
-
|
|
154
|
+
Reply only at a final state, never as mid-run progress: a text-only turn ends
|
|
155
|
+
the run, so after a compaction or phase change keep calling tools. Use one terse line:
|
|
155
156
|
|
|
156
157
|
```text
|
|
157
158
|
Autocomplete: <slug>
|
|
@@ -51,7 +51,10 @@ If the candidate diff has no matching surfaces, record
|
|
|
51
51
|
a reviewer, auditor, or test-analyst wave adds `--packet <file>`, the exact packet
|
|
52
52
|
those agents receive, and `open` refuses one that seeds a gate result, outcome,
|
|
53
53
|
count, or "already passed" claim; at Review and Seal it also needs the packet to
|
|
54
|
-
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking
|
|
54
|
+
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking through the host's own agent
|
|
55
|
+
tool (never a shell-launched host CLI such as `omp -p` or `claude -p`: that is a
|
|
56
|
+
dispatch bridge, its PID handle dies with the session, and its return is a `gap`),
|
|
57
|
+
record each returned host handle
|
|
55
58
|
(`dispatch ... start --wave <w> --role <r> --handle <id>`), then `seal` —
|
|
56
59
|
seal fails until every declared role has a distinct handle, which makes a
|
|
57
60
|
serial launch-then-wait pattern impossible to record. Record each completion
|
|
@@ -150,7 +150,8 @@ keeps chaining pending slices. One Autocomplete invocation owns all internal
|
|
|
150
150
|
backtracking. Context pressure, compaction, or a nested completion footer does
|
|
151
151
|
not create a stop; persist and resume the cursor.
|
|
152
152
|
|
|
153
|
-
|
|
153
|
+
Reply only at a final state, never as mid-run progress: a text-only turn ends
|
|
154
|
+
the run, so after a compaction or phase change keep calling tools. Use one terse line:
|
|
154
155
|
|
|
155
156
|
```text
|
|
156
157
|
Autocomplete: <slug>
|
|
@@ -51,7 +51,10 @@ If the candidate diff has no matching surfaces, record
|
|
|
51
51
|
a reviewer, auditor, or test-analyst wave adds `--packet <file>`, the exact packet
|
|
52
52
|
those agents receive, and `open` refuses one that seeds a gate result, outcome,
|
|
53
53
|
count, or "already passed" claim; at Review and Seal it also needs the packet to
|
|
54
|
-
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking
|
|
54
|
+
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking through the host's own agent
|
|
55
|
+
tool (never a shell-launched host CLI such as `omp -p` or `claude -p`: that is a
|
|
56
|
+
dispatch bridge, its PID handle dies with the session, and its return is a `gap`),
|
|
57
|
+
record each returned host handle
|
|
55
58
|
(`dispatch ... start --wave <w> --role <r> --handle <id>`), then `seal` —
|
|
56
59
|
seal fails until every declared role has a distinct handle, which makes a
|
|
57
60
|
serial launch-then-wait pattern impossible to record. Record each completion
|
|
@@ -151,7 +151,8 @@ keeps chaining pending slices. One Autocomplete invocation owns all internal
|
|
|
151
151
|
backtracking. Context pressure, compaction, or a nested completion footer does
|
|
152
152
|
not create a stop; persist and resume the cursor.
|
|
153
153
|
|
|
154
|
-
|
|
154
|
+
Reply only at a final state, never as mid-run progress: a text-only turn ends
|
|
155
|
+
the run, so after a compaction or phase change keep calling tools. Use one terse line:
|
|
155
156
|
|
|
156
157
|
```text
|
|
157
158
|
Autocomplete: <slug>
|
|
@@ -51,7 +51,10 @@ If the candidate diff has no matching surfaces, record
|
|
|
51
51
|
a reviewer, auditor, or test-analyst wave adds `--packet <file>`, the exact packet
|
|
52
52
|
those agents receive, and `open` refuses one that seeds a gate result, outcome,
|
|
53
53
|
count, or "already passed" claim; at Review and Seal it also needs the packet to
|
|
54
|
-
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking
|
|
54
|
+
name the frozen candidate diff and the manifest to cover every checkpointed path), launch every native agent nonblocking through the host's own agent
|
|
55
|
+
tool (never a shell-launched host CLI such as `omp -p` or `claude -p`: that is a
|
|
56
|
+
dispatch bridge, its PID handle dies with the session, and its return is a `gap`),
|
|
57
|
+
record each returned host handle
|
|
55
58
|
(`dispatch ... start --wave <w> --role <r> --handle <id>`), then `seal` —
|
|
56
59
|
seal fails until every declared role has a distinct handle, which makes a
|
|
57
60
|
serial launch-then-wait pattern impossible to record. Record each completion
|
|
@@ -151,7 +151,8 @@ keeps chaining pending slices. One Autocomplete invocation owns all internal
|
|
|
151
151
|
backtracking. Context pressure, compaction, or a nested completion footer does
|
|
152
152
|
not create a stop; persist and resume the cursor.
|
|
153
153
|
|
|
154
|
-
|
|
154
|
+
Reply only at a final state, never as mid-run progress: a text-only turn ends
|
|
155
|
+
the run, so after a compaction or phase change keep calling tools. Use one terse line:
|
|
155
156
|
|
|
156
157
|
```text
|
|
157
158
|
Autocomplete: <slug>
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "devrites",
|
|
3
|
-
"version": "6.9.
|
|
3
|
+
"version": "6.9.3",
|
|
4
4
|
"description": "DevRites: a disciplined senior-engineer workflow pack for Claude Code, Codex, omp, pi, and Devin CLI",
|
|
5
5
|
"license": "SEE LICENSE IN LICENSE",
|
|
6
6
|
"homepage": "https://github.com/ViktorsBaikers/DevRites#readme",
|