kushi-agents 3.13.0 → 3.14.0
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/package.json +1 -1
- package/plugin/instructions/crm-internal-vs-confirmed.instructions.md +1 -1
- package/plugin/instructions/engagement-root-resolution.instructions.md +1 -1
- package/plugin/learnings/crm.md +1 -1
- package/plugin/learnings/onenote.md +2 -2
- package/plugin/skills/pull-onenote/SKILL.md +1 -1
- package/src/settings.mjs +2 -2
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "kushi-agents",
|
|
3
|
-
"version": "3.
|
|
3
|
+
"version": "3.14.0",
|
|
4
4
|
"description": "Install Kushi — multi-source project evidence agent with snapshot+stream capture across Email, Teams, OneNote, SharePoint, Meetings, CRM, ADO. WorkIQ-only for M365 sources (Graph / m365_* FORBIDDEN as fallbacks; user-paste is first-class). Host-agnostic.",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"bin": {
|
|
@@ -8,7 +8,7 @@ applyTo: "**"
|
|
|
8
8
|
|
|
9
9
|
**A CRM field value is NEVER automatically a confirmed fact.** A Dataverse field update records an internal decision; it does not prove the decision has been (a) communicated to the customer, (b) approved by deal desk / finance / legal, or (c) confirmed in a customer-facing transcript, note, email, or meeting artifact.
|
|
10
10
|
|
|
11
|
-
Adapted into kushi from
|
|
11
|
+
Adapted into kushi from prior doctrine (see `fde-grounding.instructions.md > CRITICAL: CRM Field Values vs. Confirmed Facts`).
|
|
12
12
|
|
|
13
13
|
## Three states every CRM-sourced assertion must be classified into
|
|
14
14
|
|
|
@@ -12,7 +12,7 @@ The **engagement-root** is the parent folder that contains all of the user's eng
|
|
|
12
12
|
Resolve `<engagement-root>` in this order — first match wins:
|
|
13
13
|
|
|
14
14
|
1. **`<USER_HOME>/.copilot/project-evidence.yml`** — read `engagement_root:` field (preferred).
|
|
15
|
-
2. **`customer_workspace/FDEDocs/`** — if the user's workspace has this symlink, follow it. Common in
|
|
15
|
+
2. **`customer_workspace/FDEDocs/`** — if the user's workspace has this symlink, follow it. Common in FDE-style installs.
|
|
16
16
|
3. **Ask the user** once and persist the answer to `<USER_HOME>/.copilot/project-evidence.yml engagement_root`.
|
|
17
17
|
|
|
18
18
|
## Live config location
|
package/plugin/learnings/crm.md
CHANGED
|
@@ -14,7 +14,7 @@ A live Dataverse REST probe afternoon of 2026-05-18 — `GET /new_frontierengine
|
|
|
14
14
|
### Why this was a defect
|
|
15
15
|
The pull-crm SKILL `Resolution order` section already documented the 4-step sequence (title → account → recent-slice → ask) — but bootstrap was not actually executing it. It appears bootstrap relied on a WorkIQ-only / metadata-only probe that didn't reach Dataverse, then silently wrote `disabled: true`. That is the worst possible disposition: it pretends there is no CRM record, hides the failure from future refreshes, and the project loses CRM evidence entirely.
|
|
16
16
|
|
|
17
|
-
Adjacent observation:
|
|
17
|
+
Adjacent observation: equivalent sibling tooling performs this resolution sequence in its bootstrap and would not have missed FE-2026-001791. The doctrine was present in kushi but the execution path was weak.
|
|
18
18
|
|
|
19
19
|
### Fix
|
|
20
20
|
- New HARD-rule instruction `plugin/instructions/crm-bootstrap-discovery.instructions.md` — `disabled: true` is ONLY allowed after the FULL 4-step REST sequence returns 0 AND the user is presented with top candidates. Auth/reachability failures must leave the boundary EMPTY (with `reason: 'crm-auth-unavailable-<date>'`) so the next refresh retries — NEVER write `disabled: true`.
|
|
@@ -98,7 +98,7 @@ The first eight `pull-onenote` versions used prose phrasing (`"sectionFileId <id
|
|
|
98
98
|
|
|
99
99
|
1. WorkIQ does NOT honor `wdsectionfileid = <id>` as filter syntax — it routes to summary mode AND returns "OneNote internal properties not exposed as searchable fields" refusal text.
|
|
100
100
|
2. The wdpartid GUIDs we observed in earlier runs were **URL fragments inside SharePoint Doc.aspx hyperlinks** that WorkIQ rendered as response footnotes — not search-index extractor outputs.
|
|
101
|
-
3. The
|
|
101
|
+
3. The natural-language WorkIQ pattern (query naming the section + notebook by display name and the page by quoted title) is the actual working pattern. It returned a real verbatim body for the HCA `4/3 - HCA with Jay and Martin` page.
|
|
102
102
|
4. **Body retrieval is non-deterministic** — the same 4/3 page returned a verbatim body at 19:42 PDT and `BODY-NOT-EXPOSED` at 19:48 PDT, same query, no edits. The M365 search index's exposure of OneNote bodies oscillates over time.
|
|
103
103
|
5. **The blocker for months was the WorkIQ EULA.** Without `workiq accept-eula`, every OneNote query silently returns nothing useful. This is a one-time setup step, not a per-call gate.
|
|
104
104
|
|
|
@@ -112,7 +112,7 @@ The first eight `pull-onenote` versions used prose phrasing (`"sectionFileId <id
|
|
|
112
112
|
|
|
113
113
|
**HCA result (2026-05-14):** 18 pages enumerated. 1 captured verbatim (4/3). 15 pending retry (BODY-NOT-EXPOSED). 2 enumeration-only (will be probed in Step B on next refresh).
|
|
114
114
|
|
|
115
|
-
**Key lesson:** when a doctrine is grounded in pattern-matching against tool responses (e.g. "field names route to extractor"), validate it empirically against the live tool BEFORE shipping. The v3.7.8 doctrine was internally consistent and self-citing but never actually tested end-to-end — the 4/3 success that motivated v3.7.9 happened only after honestly retracting v3.7.8 and replicating the
|
|
115
|
+
**Key lesson:** when a doctrine is grounded in pattern-matching against tool responses (e.g. "field names route to extractor"), validate it empirically against the live tool BEFORE shipping. The v3.7.8 doctrine was internally consistent and self-citing but never actually tested end-to-end — the 4/3 success that motivated v3.7.9 happened only after honestly retracting v3.7.8 and replicating the natural-language WorkIQ workflow step-by-step.
|
|
116
116
|
|
|
117
117
|
|
|
118
118
|
## 2026-05-14 — v3.7.9 retraction + v3.8.0 architectural pivot
|
|
@@ -342,7 +342,7 @@ Used ONLY when:
|
|
|
342
342
|
- The previous run hit auth-required and retry-cooldown has not elapsed (24h), OR
|
|
343
343
|
- The user explicitly requests WorkIQ-only for diagnostic comparison.
|
|
344
344
|
|
|
345
|
-
Canonical WorkIQ Step B query (
|
|
345
|
+
Canonical WorkIQ Step B query (natural-language pattern):
|
|
346
346
|
|
|
347
347
|
> Return the FULL readable content verbatim of the page titled `<title>` in my OneNote section `<one_sectionName>` in notebook `<one_notebookName>`. Do not summarize. Do not paraphrase. If the body is not exposed, say so explicitly with the literal phrase `BODY-NOT-EXPOSED` on its own line.
|
|
348
348
|
|
package/src/settings.mjs
CHANGED
|
@@ -4,12 +4,12 @@ import { modify, applyEdits, parse, printParseErrorCode } from 'jsonc-parser';
|
|
|
4
4
|
import { SETTINGS_MAP } from './constants.mjs';
|
|
5
5
|
|
|
6
6
|
/**
|
|
7
|
-
* Merge
|
|
7
|
+
* Merge Kushi settings into .vscode/settings.json (JSONC-safe).
|
|
8
8
|
* Creates the file and directory if they don't exist.
|
|
9
9
|
* Additive-only: never removes existing entries.
|
|
10
10
|
*
|
|
11
11
|
* @param {string} projectRoot – the user's project root (cwd)
|
|
12
|
-
* @param {string} destination – relative destination path (e.g. ".
|
|
12
|
+
* @param {string} destination – relative destination path (e.g. ".kushi")
|
|
13
13
|
* @returns {{ created: boolean, keysAdded: string[], keysUnchanged: string[] }}
|
|
14
14
|
*/
|
|
15
15
|
export function mergeSettings(projectRoot, destination) {
|