pi-goal-list-loop-audit 0.34.8 → 0.34.9
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/README.md +11 -7
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -315,9 +315,17 @@ hard timeout, and the auditor aborts after 10m with zero session activity
|
|
|
315
315
|
## Recommended companions
|
|
316
316
|
|
|
317
317
|
glla is the goal plane — it drives, verifies, and notifies. It does not
|
|
318
|
-
try to be the whole rig. Four plugins round it out
|
|
319
|
-
|
|
320
|
-
|
|
318
|
+
try to be the whole rig. Four plugins round it out. The first is
|
|
319
|
+
**effectively required** — glla's drafting interviews, DECIDE findings,
|
|
320
|
+
and confirm dialogs are built around structured questions (it degrades
|
|
321
|
+
to plain-text prompts without it, but that is the fallback path, not the
|
|
322
|
+
product). The other three are optional; glla works without them:
|
|
323
|
+
|
|
324
|
+
- **`@juicesharp/rpiv-ask-user-question`** — **install this one.**
|
|
325
|
+
Structured questions with multi-select and markdown previews: the
|
|
326
|
+
/goal drafting interview, DECIDE findings, and every confirm dialog
|
|
327
|
+
render through it. Without it you get prose fallbacks — functional,
|
|
328
|
+
but not the intended UX.
|
|
321
329
|
- **`@tintinweb/pi-subagents`** — the `Agent` tool: parallel Explore /
|
|
322
330
|
Plan / general-purpose subagents. glla's prompts teach fan-out with ROI
|
|
323
331
|
(parallelize real work, never ceremony spawning) and brief discipline,
|
|
@@ -331,10 +339,6 @@ works without any of them):
|
|
|
331
339
|
judgment per call. Role clarity: the advisor is *advisory*, never
|
|
332
340
|
verification — glla's isolated auditor remains the only completion
|
|
333
341
|
gate.
|
|
334
|
-
- **`@juicesharp/rpiv-ask-user-question`** — structured questions with
|
|
335
|
-
multi-select and markdown previews. glla's drafting interviews and
|
|
336
|
-
DECIDE findings render through it when installed (plain prompts
|
|
337
|
-
otherwise).
|
|
338
342
|
- **`@pi-unipi/notify`** — push beyond the desktop: Telegram, Gotify,
|
|
339
343
|
ntfy, with per-event routing. glla's built-in pushes cover the local
|
|
340
344
|
desktop case and fire only where there is something to DO; add this
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "pi-goal-list-loop-audit",
|
|
3
|
-
"version": "0.34.
|
|
3
|
+
"version": "0.34.9",
|
|
4
4
|
"description": "Mission control for autonomous pi: interview-drafted goals, an audited task queue, and forever-loops (metric, spec, project-audit) that run for hours. An isolated extension-less auditor re-verifies every completion with raw evidence; confirmed drafts, decision pauses and consent gates keep you in charge.",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"author": "dracon",
|