Merge remote-tracking branch 'origin/main' into HEAD

Author: elkaix

.changeset/add-discussion-mode.md

@@ -0,0 +1,5 @@
+---
+"@pymodel/pythinker-code": minor
+---
+
+Add experimental Expert Talk for automatic two-model analysis, reciprocal review, and fused answers.

.changeset/refreshed-new-chat-icon.md

@@ -0,0 +1,5 @@
+---
+"@pymodel/pythinker-code": patch
+---
+
+Use a refreshed New Chat icon across desktop and mobile navigation.

AGENTS.md

@@ -127,3 +127,48 @@ Gate behind flags. Env: `PYTHINKER_CODE_EXPERIMENTAL_<NAME>` toggles one; `PYTHI
 - Hard rules that affect almost every task: update the root `AGENTS.md`.
 - Rules that only affect a specific directory: update the nearest sub-directory `AGENTS.md`.
 - Project-map entries stay at 1–2 sentences; deep package docs live in the package's own `AGENTS.md`.
+
+<!-- gitnexus:start -->
+# GitNexus — Code Intelligence
+
+This project is indexed by GitNexus as **pythinker-code** (91454 symbols, 263612 relationships, 300 execution flows). Use the GitNexus MCP tools to understand code, assess impact, and navigate safely.
+
+> Index stale? Run `node .gitnexus/run.cjs analyze` from the project root — it auto-selects an available runner. No `.gitnexus/run.cjs` yet? `npx gitnexus analyze` (npm 11 crash → `npm i -g gitnexus`; #1939).
+
+## Always Do
+
+- **MUST run impact analysis before editing any symbol.** Before modifying a function, class, or method, run `impact({target: "symbolName", direction: "upstream"})` and report the blast radius (direct callers, affected processes, risk level) to the user.
+- **MUST run `detect_changes()` before committing** to verify your changes only affect expected symbols and execution flows. For regression review, compare against the default branch: `detect_changes({scope: "compare", base_ref: "main"})`.
+- **MUST warn the user** if impact analysis returns HIGH or CRITICAL risk before proceeding with edits.
+- When exploring unfamiliar code, use `query({search_query: "concept"})` to find execution flows instead of grepping. It returns process-grouped results ranked by relevance.
+- When you need full context on a specific symbol — callers, callees, which execution flows it participates in — use `context({name: "symbolName"})`.
+- For security review, `explain({target: "fileOrSymbol"})` lists taint findings (source→sink flows; needs `analyze --pdg`).
+
+## Never Do
+
+- NEVER edit a function, class, or method without first running `impact` on it.
+- NEVER ignore HIGH or CRITICAL risk warnings from impact analysis.
+- NEVER rename symbols with find-and-replace — use `rename` which understands the call graph.
+- NEVER commit changes without running `detect_changes()` to check affected scope.
+
+## Resources
+
+| Resource | Use for |
+|----------|---------|
+| `gitnexus://repo/pythinker-code/context` | Codebase overview, check index freshness |
+| `gitnexus://repo/pythinker-code/clusters` | All functional areas |
+| `gitnexus://repo/pythinker-code/processes` | All execution flows |
+| `gitnexus://repo/pythinker-code/process/{name}` | Step-by-step execution trace |
+
+## CLI
+
+| Task | Read this skill file |
+|------|---------------------|
+| Understand architecture / "How does X work?" | `.claude/skills/gitnexus/gitnexus-exploring/SKILL.md` |
+| Blast radius / "What breaks if I change X?" | `.claude/skills/gitnexus/gitnexus-impact-analysis/SKILL.md` |
+| Trace bugs / "Why is X failing?" | `.claude/skills/gitnexus/gitnexus-debugging/SKILL.md` |
+| Rename / extract / split / refactor | `.claude/skills/gitnexus/gitnexus-refactoring/SKILL.md` |
+| Tools, resources, schema reference | `.claude/skills/gitnexus/gitnexus-guide/SKILL.md` |
+| Index, status, clean, wiki CLI commands | `.claude/skills/gitnexus/gitnexus-cli/SKILL.md` |
+
+<!-- gitnexus:end -->

CLAUDE.md

@@ -1 +1,46 @@
 @AGENTS.md
+
+<!-- gitnexus:start -->
+# GitNexus — Code Intelligence
+
+This project is indexed by GitNexus as **pythinker-code** (91454 symbols, 263612 relationships, 300 execution flows). Use the GitNexus MCP tools to understand code, assess impact, and navigate safely.
+
+> Index stale? Run `node .gitnexus/run.cjs analyze` from the project root — it auto-selects an available runner. No `.gitnexus/run.cjs` yet? `npx gitnexus analyze` (npm 11 crash → `npm i -g gitnexus`; #1939).
+
+## Always Do
+
+- **MUST run impact analysis before editing any symbol.** Before modifying a function, class, or method, run `impact({target: "symbolName", direction: "upstream"})` and report the blast radius (direct callers, affected processes, risk level) to the user.
+- **MUST run `detect_changes()` before committing** to verify your changes only affect expected symbols and execution flows. For regression review, compare against the default branch: `detect_changes({scope: "compare", base_ref: "main"})`.
+- **MUST warn the user** if impact analysis returns HIGH or CRITICAL risk before proceeding with edits.
+- When exploring unfamiliar code, use `query({search_query: "concept"})` to find execution flows instead of grepping. It returns process-grouped results ranked by relevance.
+- When you need full context on a specific symbol — callers, callees, which execution flows it participates in — use `context({name: "symbolName"})`.
+- For security review, `explain({target: "fileOrSymbol"})` lists taint findings (source→sink flows; needs `analyze --pdg`).
+
+## Never Do
+
+- NEVER edit a function, class, or method without first running `impact` on it.
+- NEVER ignore HIGH or CRITICAL risk warnings from impact analysis.
+- NEVER rename symbols with find-and-replace — use `rename` which understands the call graph.
+- NEVER commit changes without running `detect_changes()` to check affected scope.
+
+## Resources
+
+| Resource | Use for |
+|----------|---------|
+| `gitnexus://repo/pythinker-code/context` | Codebase overview, check index freshness |
+| `gitnexus://repo/pythinker-code/clusters` | All functional areas |
+| `gitnexus://repo/pythinker-code/processes` | All execution flows |
+| `gitnexus://repo/pythinker-code/process/{name}` | Step-by-step execution trace |
+
+## CLI
+
+| Task | Read this skill file |
+|------|---------------------|
+| Understand architecture / "How does X work?" | `.claude/skills/gitnexus/gitnexus-exploring/SKILL.md` |
+| Blast radius / "What breaks if I change X?" | `.claude/skills/gitnexus/gitnexus-impact-analysis/SKILL.md` |
+| Trace bugs / "Why is X failing?" | `.claude/skills/gitnexus/gitnexus-debugging/SKILL.md` |
+| Rename / extract / split / refactor | `.claude/skills/gitnexus/gitnexus-refactoring/SKILL.md` |
+| Tools, resources, schema reference | `.claude/skills/gitnexus/gitnexus-guide/SKILL.md` |
+| Index, status, clean, wiki CLI commands | `.claude/skills/gitnexus/gitnexus-cli/SKILL.md` |
+
+<!-- gitnexus:end -->

CONTEXT.md

@@ -0,0 +1,69 @@
+# Pythinker Code
+
+Pythinker Code is a provider-agnostic AI coding agent for terminal, browser, desktop, and editor workflows.
+
+## Language
+
+**Expert Talk**:
+A structured conversation between two user-selected AI models that returns one fused answer.
+_Avoid_: Debate mode, fusion mode
+
+**Expert Talk pair**:
+The two ordered configured model IDs saved for future Expert Talk runs.
+_Avoid_: Global pair, secondary model
+
+**Configured model ID**:
+The stable catalog key that identifies one configured model record and can be stored in an Expert Talk pair.
+_Avoid_: Display name, provider model name, loose alias
+
+**Effective model binding**:
+The immutable execution identity produced for one Expert Talk role after current catalog and routing validation.
+_Avoid_: Selected model, live alias
+
+**Pair collapse**:
+An Expert Talk admission failure where two configured selections resolve to the same effective execution target.
+_Avoid_: Duplicate model choice
+
+**Expert Talk admission**:
+The pre-acceptance validation that resolves and freezes an eligible, distinct pair for one run.
+_Avoid_: Provider ping, opening stage
+
+**Fusion Lead**:
+The first model in an Expert Talk pair. It participates as an equal expert, then its immutable model binding supplies a fresh inference for fusion.
+_Avoid_: Architect, judge, Model 1
+
+**Peer Expert**:
+The second model in an Expert Talk pair. It participates as an equal expert in opening and peer review but does not own fusion.
+_Avoid_: Secondary model, Model 2
+
+**Expert Talk activation**:
+The transient intent to use Expert Talk for the next accepted user turn.
+_Avoid_: Expert Talk mode, persistent activation
+
+**Expert Talk run**:
+One recorded Expert Talk execution owned by one accepted user turn.
+_Avoid_: Resumed run, shared run
+
+**Expert Talk exchange**:
+Two independent openings followed by one symmetric peer review between the run-scoped Fusion Lead and Peer Expert.
+_Avoid_: Debate, adaptive conversation
+
+**Fused answer**:
+The single direct response produced by a fresh Fusion Lead inference after the Expert Talk exchange.
+_Avoid_: Verdict, winning answer
+
+**Fusion notes**:
+Structured consensus, divergence, uncertainty, and attribution stored with an Expert Talk run but excluded from the normal assistant answer.
+_Avoid_: Diagnostic appendix, debate transcript
+
+**Expert Talk stage artifact**:
+An immutable complete or partial opening, review, or fusion output owned by one role and stage and kept outside the normal transcript.
+_Avoid_: Participant message, hidden memory
+
+**Expert Talk result**:
+The versioned successful-run record that contains the fused answer and separate fusion notes.
+_Avoid_: Meeting transcript, winning opinion
+
+**Expert Talk retry**:
+A new run for the same accepted turn input that reruns admission and every stage without resuming participants or reusing artifacts.
+_Avoid_: Resume, fusion-only retry

apps/pythinker-code/dist-web/.web-bundle-manifest.json

@@ -1,4 +1,4 @@
 {
-  "sourceHash": "f7cf5febcdf56552c8623ab242e49ecff60f2d3bb0a307d139a04c7225217ed1",
-  "sourceFileCount": 412
+  "sourceHash": "0a37c819db91d9c13c59e2b4cd1b26e5e6eda54d5dab911b7d233bc3e6a1868a",
+  "sourceFileCount": 417
 }