@abelxiaoxing/cadence 1.5.0 → 1.5.1
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
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@abelxiaoxing/cadence",
|
|
3
|
-
"version": "1.5.
|
|
3
|
+
"version": "1.5.1",
|
|
4
4
|
"description": "Four explicit Abel workflow prompts with stage-isolated design and diagnosis, durable resumable implementation, and three package-owned professional Agents.",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"type": "module",
|
package/prompts/abel-design.md
CHANGED
|
@@ -252,7 +252,6 @@ Do not expand write sets merely to satisfy a disposition label.
|
|
|
252
252
|
For public UI or API work, seal the relevant route authorization, page state, API response and contract, public HTML/template/theme behavior, and an approved browser E2E check when those surfaces are affected.
|
|
253
253
|
|
|
254
254
|
The plan admits only shell-free `vitest`, `package-script`, `static-check`, or ordered `steps` contracts.
|
|
255
|
-
`dev-browser` is required only by an approved browser E2E contract; its absence does not block unrelated tasks or stages.
|
|
256
255
|
Pin local runners, package scripts, arguments, `minTests`, and `noInstall` behavior; sealed classifications are derived from phase/purpose when omitted.
|
|
257
256
|
Reject shell operators, implicit downloads, absolute/escaping paths, missing local capability, and unsupported verification shapes during Design readiness.
|
|
258
257
|
|
package/prompts/abel-diagnose.md
CHANGED
|
@@ -57,7 +57,7 @@ Do not invent a repair.
|
|
|
57
57
|
If the requested result actually requires new observable behavior, a new dependency, wider paths, substantive architecture/policy, or another product decision, return `scope-decision-required` with the evidence and decision that the user must make.
|
|
58
58
|
Do not silently widen Diagnose and do not transform an ordinary repair failure into that result.
|
|
59
59
|
|
|
60
|
-
When an approved browser E2E check
|
|
60
|
+
When an approved browser E2E check cannot run, pause only that check and provide executable remediation; continue independent non-browser evidence when safe.
|
|
61
61
|
|
|
62
62
|
Finish only when reproduction, falsification, failing regression, minimum repair, affected verification, and baseline comparison are all evidenced.
|
|
63
63
|
After presenting the final structural diagnosis/repair result, send `{"action":"finish"}` so private dispatch is deactivated.
|
|
@@ -204,7 +204,7 @@ Runtime/session ids, credentials, approval state, and dirty-state ledgers never
|
|
|
204
204
|
Task checkboxes are parent-owned progress facts.
|
|
205
205
|
The loader may normalize only `[x]`/`[X]` to `[ ]` for the sealed task ids when checking the delivery hash; every other byte remains integrity-bound.
|
|
206
206
|
|
|
207
|
-
When an approved browser E2E check
|
|
207
|
+
When an approved browser E2E check cannot run, pause only that check; continue independent work when safe.
|
|
208
208
|
|
|
209
209
|
## Truthful completion
|
|
210
210
|
|
package/prompts/abel-init.md
CHANGED
|
@@ -43,7 +43,7 @@ Resolve and report the canonical target root before action.
|
|
|
43
43
|
|
|
44
44
|
Discover the bundled `context7-auto-research` and `grok-search` Skills and report their resolved paths.
|
|
45
45
|
Their absence does not prevent OpenSpec or AGENTS repair, but final readiness is `partial` and must include an actionable package-resource restore command.
|
|
46
|
-
Do not probe `git-commit
|
|
46
|
+
Do not probe `git-commit` or a dedicated time Skill; none is an Init prerequisite.
|
|
47
47
|
|
|
48
48
|
Report one final result with the root, selected toolchain, actions actually taken, OpenSpec capability/schema/template evidence, AGENTS files changed or unchanged, research-Skill paths or remediation, and `ready | partial | paused`.
|
|
49
49
|
Never report success from the initial probe when the post-write recheck did not pass.
|