wicked-garden 12.50.0 → 12.51.0
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/.claude-plugin/marketplace.json +1 -1
- package/.claude-plugin/plugin.json +1 -1
- package/CHANGELOG.md +12 -0
- package/package.json +1 -1
- package/skills/qe/refs/setup.md +12 -4
|
@@ -1 +1 @@
|
|
|
1
|
-
{"name": "wicked-garden", "owner": {"name": "Mike Parcewski"}, "plugins": [{"name": "wicked-garden", "source": "./", "description": "Your coding agent already plans and swarms \u2014 Wicked Garden is the curated toolkit for what it can't: proof-not-claims gates, code links grep can't see, deterministic refactor, cross-session memory, multi-model second opinions (fallback seats labelled). Don't fight the harness; fill its gaps. The gate needs wicked-vault (a self-contained infra peer, installed directly) and re-derives through garden's built-in engine; estate/bus are opt-in layers (wicked-testing and wicked-brain are retired \u2014 QE ships in-catalog as the qe domain, memory lives in wicked-estate).", "version": "12.
|
|
1
|
+
{"name": "wicked-garden", "owner": {"name": "Mike Parcewski"}, "plugins": [{"name": "wicked-garden", "source": "./", "description": "Your coding agent already plans and swarms \u2014 Wicked Garden is the curated toolkit for what it can't: proof-not-claims gates, code links grep can't see, deterministic refactor, cross-session memory, multi-model second opinions (fallback seats labelled). Don't fight the harness; fill its gaps. The gate needs wicked-vault (a self-contained infra peer, installed directly) and re-derives through garden's built-in engine; estate/bus are opt-in layers (wicked-testing and wicked-brain are retired \u2014 QE ships in-catalog as the qe domain, memory lives in wicked-estate).", "version": "12.51.0"}], "version": "12.51.0"}
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "wicked-garden",
|
|
3
|
-
"version": "12.
|
|
3
|
+
"version": "12.51.0",
|
|
4
4
|
"description": "Your coding agent's harness already plans and swarms; Wicked Garden is the curated toolkit for what it can't do alone. Proof, not claims \u2014 gates re-derive 'done' through the internal loom engine (fail-closed, independently attested via wicked-vault), so a self-asserted pass can't lie its way green. Relationships grep can't see \u2014 the wicked-estate code graph + injected bus/dispatch/capability/archetype edges power blast-radius & lineage. Plus deterministic multi-file refactor (wicked-patch), cross-session memory + knowledge (the mem domain over wicked-estate), multi-model second opinions (fallback seats labelled) (jam:council), portable expertise as on-demand skill-refs, and evidence-gated testing (the in-catalog qe domain). It reads the shape of your work to apply the right rigor \u2014 steering, not a pipeline \u2014 then gets out of the way. The compiler stamps a self-contained evidence gate into any repo that runs with no wicked-garden installed. Don't fight the harness; fill its gaps. The evidence gate requires wicked-vault, a self-contained infra peer wicked-garden installs directly; wicked-bus is an opt-in layer \u2014 install what you'll use, the toolkit works without the rest.",
|
|
5
5
|
"wicked_vault_version": "^0.5.0",
|
|
6
6
|
"wicked_ledger_version": "^0.6.0",
|
package/CHANGELOG.md
CHANGED
|
@@ -2,6 +2,12 @@
|
|
|
2
2
|
|
|
3
3
|
## [Unreleased]
|
|
4
4
|
|
|
5
|
+
## [12.51.0] — 2026-10-11
|
|
6
|
+
|
|
7
|
+
### Fixed
|
|
8
|
+
- **`wicked-garden-qe` setup scaffolds the ledger root with Python (wicked-core#881, wicked-crew#951, PR #1277).** It no longer uses a shell `mkdir` of `$WICKED_QE_LEDGER_DIR`. A governed run's write fence judges a shell write target as written, and the unexpanded variable read as a path inside the worktree, so a governed QE unit's scaffold was refused (rig run 46fb2c04). With wicked-core#889 (core-ts 0.9.2), the QE unit runs garden's pipeline into its own roots.
|
|
9
|
+
|
|
10
|
+
|
|
5
11
|
## [12.50.0] — 2026-10-11
|
|
6
12
|
|
|
7
13
|
### Added
|
|
@@ -13,6 +19,12 @@
|
|
|
13
19
|
|
|
14
20
|
### Changed
|
|
15
21
|
- **The estate client follows estate's paging cursor to the whole answer (estate#279, PR #1272).**
|
|
22
|
+
- **Memory creation keeps learnings, not narration (w9-memfix, PR #1274).**
|
|
23
|
+
- The Stop hook's regex fact extraction and its auto-memorize drain are removed: `scripts/mem/session_fact_extractor.py`, `scripts/mem/auto_memorize.py`, and the `wicked.garden.fact.extracted` event. They filed about 333 sentence fragments as semantic facts.
|
|
24
|
+
- `pre_compact` saves a working memory only when tasks are in progress, always scoped `project:<cwd-basename>` (it wrote one empty, unscoped "WIP snapshot" per compaction).
|
|
25
|
+
- Memory recall (`recall`, `sources`, the hooks' context backend) defaults to the current project and its ancestors; every scope is the explicit `scope_prefix: ""` opt-in. `recall`/`sources` pass the repo intent, so faceted repo-learn memories are reachable, and `sources` labels every item `memory · <kind> · <date> · <scope>` or `knowledge · src <source>`.
|
|
26
|
+
- `store` / `capture-batch` refuse non-durable content (SHAs, versions, PR/run ids, PIDs, "today", first-person narration, under 40 chars) and source-cited claims (knowledge: use `write`). mem-capture and mem store apply the durability and told/witnessed/decided tests; the event → episode row is gone.
|
|
27
|
+
- Shim proposals need `facets.project`; repo-learn proposes code facts as `kind_type: "knowledge"` (`content, class, source, source_rev`) and never graph metrics.
|
|
16
28
|
|
|
17
29
|
## [12.49.0] — 2026-10-11
|
|
18
30
|
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "wicked-garden",
|
|
3
|
-
"version": "12.
|
|
3
|
+
"version": "12.51.0",
|
|
4
4
|
"type": "module",
|
|
5
5
|
"description": "Curated toolkit for what AI coding agents can't do alone: evidence-gated work, codegraph relationships, deterministic multi-file refactor, multi-model second opinions, and cross-session memory. Installs as a Claude Code plugin.",
|
|
6
6
|
"keywords": [
|
package/skills/qe/refs/setup.md
CHANGED
|
@@ -49,11 +49,19 @@ npx --no-install cypress --version > /dev/null 2>&1 && echo "npx-cypress: true"
|
|
|
49
49
|
The ledger root is `$WICKED_QE_LEDGER_DIR` when it is set (a governed crew run's own ledger,
|
|
50
50
|
outside the worktree), else `.wicked-qe`:
|
|
51
51
|
|
|
52
|
+
Create it with Python, not a shell `mkdir` of the variable: a governed run's write fence
|
|
53
|
+
judges a shell write target as written, and an unexpanded `"$WICKED_QE_LEDGER_DIR/..."` reads
|
|
54
|
+
as a path inside the worktree, so it is refused.
|
|
55
|
+
|
|
52
56
|
```bash
|
|
53
|
-
|
|
54
|
-
|
|
55
|
-
|
|
56
|
-
|
|
57
|
+
PY="$(command -v python3 || command -v python)"
|
|
58
|
+
"$PY" -c "
|
|
59
|
+
import os
|
|
60
|
+
root = os.environ.get('WICKED_QE_LEDGER_DIR') or '.wicked-qe'
|
|
61
|
+
for d in ('projects', 'strategies', 'scenarios', 'runs', 'verdicts', 'tasks', 'evidence'):
|
|
62
|
+
os.makedirs(os.path.join(root, d), exist_ok=True)
|
|
63
|
+
print(root)
|
|
64
|
+
"
|
|
57
65
|
```
|
|
58
66
|
|
|
59
67
|
### 3. Write config.json
|