@llblab/pi-state-flow 0.1.4 → 0.1.5

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 CHANGED
@@ -1,5 +1,11 @@
1
1
  # Changelog
2
2
 
3
+ ## 0.1.5: Skill attribution and state model clarification
4
+
5
+ - Attribute successful Skill acquisition to finalized `tool_execution_start` arguments, with a tested compatibility fallback to the intercepted `tool_call` input.
6
+ - Clarify that the three-field materialized handoff coexists with mutable exogenous project and runtime state that can also shape model behavior.
7
+ - Remove the obsolete `/reload` step from the usage instructions.
8
+
3
9
  ## 0.1.4: Release verification hotfix
4
10
 
5
11
  - Kept npm pack lifecycle output out of release inventory JSON so Trusted Publisher releases can complete public verification and GitHub Release creation.
package/README.md CHANGED
@@ -32,7 +32,6 @@ pi install git:github.com/llblab/pi-state-flow
32
32
  ## Usage
33
33
 
34
34
  ```text
35
- /reload
36
35
  /state-flow-start
37
36
  ```
38
37
 
@@ -90,6 +89,8 @@ This removes `working.move`, replaces `working.result`, preserves every other me
90
89
 
91
90
  State and patch size have no byte, growth, pressure, or project-schema limit. Patch history remains only in the Pi trace, not in model context.
92
91
 
92
+ These three fields are the complete **materialized State Flow state**, but they are not the only state that can shape model behavior. A situational fourth, exogenous state lives outside the handoff: the current project, workspace, tools, processes, and runtime environment. It can change while work is in progress—including through the model's own tool effects—and later observations of those changes can alter subsequent behavior. State Flow neither snapshots nor rolls back this external state; `working` should retain only the decision-relevant facts needed to reconnect the next run to it.
93
+
93
94
  ## Terminal handoff quality
94
95
 
95
96
  The terminal patch is a required handoff, not a progress phrase or transcript summary. The next run may see only this state plus its new user prompt. A fresh model should be able to continue without rereading, rediscovering, re-deriving decisions, or repeating failed approaches.
package/index.ts CHANGED
@@ -475,6 +475,12 @@ export default function stateFlowExtension(pi: ExtensionAPI): void {
475
475
  pendingSkillReads.set(event.toolCallId, { toolName: event.toolName, args: event.input });
476
476
  });
477
477
 
478
+ pi.on("tool_execution_start", (event) => {
479
+ const pending = pendingSkillReads.get(event.toolCallId);
480
+ if (!snapshot.enabled || !pending) return;
481
+ pendingSkillReads.set(event.toolCallId, { toolName: event.toolName, args: event.args });
482
+ });
483
+
478
484
  pi.on("tool_execution_end", (event) => {
479
485
  const pending = pendingSkillReads.get(event.toolCallId);
480
486
  pendingSkillReads.delete(event.toolCallId);
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@llblab/pi-state-flow",
3
- "version": "0.1.4",
3
+ "version": "0.1.5",
4
4
  "private": false,
5
5
  "description": "State Flow terminal-handoff runtime for Pi, inspired by SKILL.state",
6
6
  "keywords": ["pi-package", "pi-extension", "state-management"],