@leing2021/super-pi 0.29.0 → 0.30.0

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 CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@leing2021/super-pi",
3
- "version": "0.29.0",
3
+ "version": "0.30.0",
4
4
  "private": false,
5
5
  "type": "module",
6
6
  "description": "Pi-native Compound Engineering package for iterative development workflows",
@@ -25,7 +25,7 @@ See [shared pipeline instructions](../references/pipeline-config.md) for model r
25
25
  - Extract keywords → `grep -rl "tags:.*keyword" docs/solutions/ ~/.pi/agent/docs/solutions/`
26
26
  - Read **frontmatter** only (first 15 lines) of matches → score by severity + tag relevance
27
27
  - Fully read top 3 candidates
28
- 8. **Spec axis:** when a plan artifact exists, compare diff against it — report **missing** requirements, **scope creep** (unrequested behaviour), and **wrong implementation** (looks done but isn't). Also **trace back** to the user's original wording (brainstorm scope) to catch directional misunderstandings the plan itself encoded. Skip if no plan.
28
+ 8. **Spec axis:** determine spec source in priority order — (a) plan artifact; (b) brainstorm artifact (trace back to original wording to catch directional misunderstandings the plan encoded); (c) issue references in commit messages (`git log <base>..HEAD --oneline`, scan for `#123` / `Closes #45` / `!67`) identify the ref and ask the user whether to treat it as spec source, do **not** auto-fetch; (d) skip if none. Against the chosen spec, report **missing** requirements, **scope creep** (unrequested behaviour), and **wrong implementation** (looks done but isn't).
29
29
  9. Produce structured findings using `references/findings-schema.md`
30
30
  10. **Autofixable findings:** apply and re-review (max 3 iterations)
31
31
 
@@ -59,9 +59,9 @@ Code review is **technical evaluation**, not social performance:
59
59
  ## Workflow
60
60
 
61
61
  1. **Load context**: consume latest handoff before any broad file reads — `context_handoff load` or read `.context/compound-engineering/handoffs/latest.md`. If found, use `activeFiles`, `artifacts.plan` as starting point. If not found, proceed normally. Read `CONTEXT.md` if it exists at root — see `../references/domain-language.md`.
62
- 2. Determine diff scope from branch or explicit target
62
+ 2. Determine diff scope prefer `branch`/`base` from latest handoff if present; else from explicit target; else ask user
63
63
  3. Collect stats (files, insertions, deletions) → call `review_router`
64
- 4. Read matching plan artifact
64
+ 4. Read matching plan artifact. If absent, scan commit messages for issue refs (`git log <base>..HEAD --oneline`) and ask user whether to use as spec source — do not auto-fetch
65
65
  5. Run solution search
66
66
  6. Apply each reviewer persona from `review_router`
67
67
  7. Merge into structured findings
@@ -109,6 +109,10 @@ When a stage produces or updates handoff-lite, use this evidence-first structure
109
109
  ## Recently Accessed Files
110
110
  - files recently read or edited
111
111
 
112
+ ## Git Context
113
+ - branch: <current branch or N/A>
114
+ - base: <diff base ref, e.g. main / N/A>
115
+
112
116
  ## Artifacts
113
117
  - requirements: <path or N/A>
114
118
  - plan: <path or N/A>