@curdx/flow 7.1.12 → 7.1.14
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 +25 -0
- package/package.json +1 -1
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,31 @@
|
|
|
2
2
|
|
|
3
3
|
All notable changes to `@curdx/flow` are documented here. Format follows [Keep a Changelog](https://keepachangelog.com/) and the project follows [Semantic Versioning](https://semver.org/).
|
|
4
4
|
|
|
5
|
+
## 7.1.14 — 2026-05-10
|
|
6
|
+
|
|
7
|
+
### Fixed
|
|
8
|
+
|
|
9
|
+
- **Fresh spec state now consistently uses the bounded `maxGlobalIterations: 30` default.** `/curdx-flow:start`, `/curdx-flow:new`, quick mode, and the execution-state hook fallback now agree with the implement skill's cost-runaway contract. Existing legacy state files that already store `100` remain compatible.
|
|
10
|
+
- **Agent Teams are no longer documented as the stable default anywhere in the workflow path.** Research, triage, scanner resume guidance, and core delegation rules now use direct `Task(...)` dispatch as the baseline, with Agent Teams only as an optional experimental overlay when enabled and available.
|
|
11
|
+
- **Public entrypoint skill tool grants are narrower.** Shipped `/curdx-flow:*` skills no longer use `allowed-tools: "*"`, reducing accidental broad permission grants while preserving the tools each workflow actually needs.
|
|
12
|
+
- **Support skills are hidden from the slash menu but remain model-invocable background guidance.** `spec-workflow`, `curdx-core`, `communication-style`, `interview-framework`, and `verification-before-completion` now have trigger-focused descriptions and explicit background-skill metadata.
|
|
13
|
+
- **Help and cost references now expose the current iteration-cap interface.** `/curdx-flow:help` documents both `--max-task-iterations` and `--max-global-iterations`; cost docs use the real `/curdx-flow:implement` invocation form.
|
|
14
|
+
|
|
15
|
+
### Added
|
|
16
|
+
|
|
17
|
+
- **Runner guards for runtime-contract drift.** Manifest integrity tests now lock bounded iteration defaults, forbid wildcard grants on public entrypoints, require support skills to stay hidden/background-invocable, reject old "research team as default" wording, and verify help documents the global iteration cap.
|
|
18
|
+
|
|
19
|
+
## 7.1.13 — 2026-05-10
|
|
20
|
+
|
|
21
|
+
### Changed
|
|
22
|
+
|
|
23
|
+
- **Public skill descriptions are now trigger-focused.** All 15 `/curdx-flow:*` entrypoint skills now use concise `Use when...` descriptions so Claude Code surfaces user situations instead of workflow summaries.
|
|
24
|
+
- **Skill maintenance policy captures the adopted reference-project patterns.** Added `skills/spec-workflow/references/skill-quality-patterns.md`, distilling the useful parts of the official Claude Code skills docs, `gstack`, and `superpowers`: trigger-only descriptions, deliberate manual entrypoints, progressive disclosure, coordinator posture, and verification gates.
|
|
25
|
+
|
|
26
|
+
### Added
|
|
27
|
+
|
|
28
|
+
- **Runner guard for skill description quality.** Manifest integrity tests now require public entrypoint descriptions to be trigger-focused and terse, and assert the new skill-quality reference is linked from `spec-workflow`.
|
|
29
|
+
|
|
5
30
|
## 7.1.12 — 2026-05-10
|
|
6
31
|
|
|
7
32
|
### Fixed
|