Commit Graph

319 Commits

Author SHA1 Message Date
Bill
023809ef4b Sprint 20-25 architecture plans — completing the roadmap
Sprint 20: Legacy code ingestion + modernization workflows
Sprint 21: Semantic cross-language transpilation engine
Sprint 22: Assembly languages (x86/ARM) + C++ remaining gaps
Sprint 23: Architect mode — problem description to tech stack + skeleton
Sprint 24: Security annotations, OWASP detection, secure transpilation
Sprint 25: Self-hosting, end-to-end scenarios, polish, release prep

Cumulative: ~508 steps, ~5000 tests, 19+ languages, 90+ MCP tools,
80+ annotation types across 10 subjects. Post-25 training data harvest
builds on real workflow decisions accumulated across all sprints.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-14 05:19:22 +00:00
Bill
4f5542a47f Sprint 12-19 architecture plans + roadmap through Sprint 25
Revised Sprint 11e from training data export to workflow annotation
foundation (routing types, skeleton AST, inference-to-routing bridge).
Detailed plans for Sprints 12-19 covering workflow model, GUI overhaul,
6 new languages, orchestration engine, C++ self-hosting, Claude Code
plugin, and workflow visualization. Created ARCHITECT.md with core
thesis and design principles. Training data deferred to post-25 where
real workflow decisions provide superior signal.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-14 05:14:45 +00:00
Bill
43ccafbdc7 Sprint 11a: complete steps 295-296 semanno sidecar + integration tests 2026-02-13 19:42:28 +00:00
Bill
d4a3609050 Steps 290-294: Phase 11a — Semanno format, annotation codegen, visitor dispatch (60/60 tests)
Semanno comment format standard (@semanno:type(key=value)) with emitter/parser
covering all 67+ annotation types. SemannoAnnotationImpl CRTP mixin provides
default visitor implementations for all 56 extended annotation methods across
7 language generators. Virtual inheritance resolves diamond ambiguity between
ProjectionGenerator and SemannoAnnotationImpl.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-13 18:41:31 +00:00
Bill
8cbeef5af4 Steps 284-289: Phase 10e — environment layer, host boundary nodes, MCP env tools (200/200 tests)
Completes Sprint 10. Adds EnvironmentSpec AST node, capability validation
(E0501), annotation compatibility checks (E0502-E0505), env-aware lowering
hints, 3 host boundary AST nodes (HostCall, ScheduleTask, ModuleLoad),
and 4 MCP environment tools. 38 MCP tools total.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-13 05:56:02 +00:00
Bill
ed8618e49f Phase 10e WIP: environment layer core + host boundary nodes
EnvironmentSpec, CapabilityRequirement, HostCall, ScheduleTask, ModuleLoad
classes created. All wired through Serialization.h, CompactAST.h,
SidecarPersistence.h, HeadlessAgentRPCHandler.h (setEnvironment,
getEnvironment, validateEnvironment, getLoweringHints RPCs).
Tests for steps 284-285 written. Steps 286-289 tests remaining.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:41:16 +00:00
Bill
090320f165 Steps 278-283: Phase 10d — shims, optimization, meta-programming & policy annotations (117/117 tests)
- Step 278: Shim/escape hatch — Intrinsic, Raw, CallingConv, Link, Shim, PointerArithmetic, Opaque (25 tests)
- Step 279: Platform/provenance — Target, Feature, Original, Mapping (19 tests)
- Step 280: Optimization completion — TailCall, Loop, Data, Align, Pack, BoundsCheck, Overflow (22 tests)
- Step 281: Meta-programming — Meta, Symbol, Evaluate, Template, Synthetic (20 tests)
- Step 282: Strategy/policy — Policy, Ambiguity, Candidate, Tradeoff, Choice, Decision (22 tests)
- Step 283: Integration tests — FFI workflow, sidecar roundtrip, RPC, taxonomy completeness (9 tests)
- 29 new annotation classes with JSON roundtrip, compact AST, sidecar persistence
- All 58 semantic annotation types recognized across 8 subjects

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:29:43 +00:00
Bill
e2d1872f35 Steps 272-277: Phase 10c — type system, concurrency & scope annotations (68/68 tests)
- Step 272: Type layout/constraints — BitWidth, Endian, Layout, Nullability, Variance (37 tests)
- Step 273: Type identity/mutability — Identity, Mut, TypeState (25 tests)
- Step 274: Concurrency primitives — Atomic, Sync, ThreadModel, MemoryBarrier (25 tests)
- Step 275: Async/error handling — Exec, Blocking, Parallel, Trap, Exception, Panic (36 tests)
- Step 276: Scope/namespace — Binding, Lookup, Capture, Visibility, Namespace, Scope (32 tests)
- Step 277: Integration tests — cross-subject workflows, sidecar roundtrip, RPC (16 tests)
- 24 new annotation classes with JSON roundtrip, compact AST, sidecar persistence
- Generic fallback in getSemanticAnnotations RPC for extensible annotation queries

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 18:15:21 +00:00
Bill
a049d6010b Steps 269-271: Phase 10b — annotation agent interface (32/32 tests)
- Step 269: RPC annotation methods (set/get/remove/unannotated) with permission enforcement (12 tests)
- Step 270: MCP prompt templates for annotation workflows (12 tests)
- Step 271: Integration tests for full agent annotation pipeline (8 tests)
- Fixed sideEffectHint heuristic to check descendants of body statements

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 16:18:25 +00:00
Bill
976161dc4a Steps 266-268: Phase 10a — semantic annotation core + sidecar persistence (32/32 tests)
Step 266: 5 semantic annotation AST types (Intent, Complexity, Risk,
Contract, SemanticTag) with JSON roundtrip and compact AST integration.
Step 267: Sidecar AST persistence (.whetstone/<file>.ast.json) with
save/load/list RPC methods, MCP tools, and permission enforcement.
Step 268: Phase 10a integration tests — multi-file sidecar workflow,
all 5 types in compact view, idempotent roundtrip, source isolation.

Also includes Sprint 10 plan, GUI/installer polish, and Emacs integration fixes.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 16:01:56 +00:00
Bill
c36fd92045 Steps 263-265: Phase 9e — save buffer to disk, undo/redo, integration tests
Step 263: saveBuffer/saveAllBuffers RPC methods write editBuf to disk with
workspace path resolution and parent dir creation. 12/12 tests.

Step 264: State-based undo/redo via HeadlessUndoStack. Snapshots recorded
automatically before mutations; undo/redo restore AST + editBuf. 12/12 tests.

Step 265: Phase 9e integration tests — save after undo/redo, mixed undo
states across buffers, full open→mutate→diagnose→undo→redo→save. 8/8 tests.

Phase 9e complete: 32/32 tests. Sprint 9 total: 244/244 across 5 phases.
34 MCP tools.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 01:27:23 +00:00
Bill
872ef493d2 Step 262: Phase 9d multi-file project integration tests (Phase 9d complete)
End-to-end tests for the full multi-file workflow: open 3 files, switch
buffers, cross-file symbols, import graph, project diagnostics, search,
rename across files, and batch query verification. 8/8 tests passing.
Sprint 9 complete: 212/212 tests across all 4 phases (Steps 245-262).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-12 00:46:59 +00:00
Bill
eab979bae3 Step 261: project-wide search and refactor (searchProject, renameSymbol)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 20:27:47 +00:00
Bill
8a461df47e Step 260: project-wide diagnostics (getProjectDiagnostics)
Diagnostics across all open files in one call, grouped by file path.
Includes cross-file E0400 warnings for undefined imports, with optional
severity filter and file glob filter. 12/12 tests pass.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 20:13:46 +00:00
Bill
f0417bccce Step 259: cross-file symbol resolution and import graph
getInScopeSymbols gains crossFile parameter to include exported symbols from
all open buffers. Import graph tracks file dependencies, auto-maintained on
openFile/closeFile. Text-based fallback for parsers without Import AST nodes.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 20:06:46 +00:00
Bill
ac5b7e21c8 Step 258: project model and workspace indexing (Phase 9d start)
Adds multi-buffer management via openFile/closeFile/listBuffers/setActiveBuffer
RPC methods, plus indexWorkspace for scanning workspace files without opening
them. Language auto-detected from extension. 27 MCP tools total (was 22).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 19:43:53 +00:00
Bill
5830361463 Step 257: token efficiency tests + benchmarks (Phase 9c complete)
Systematic validation of all Phase 9c token-saving features: compact AST
(6-7% of full size), lean scope (4% of detailed), diagnostic deltas, budget
pagination across 3 pages, batch queries, and combined 95% savings. Benchmark
confirms sub-millisecond parse->mutate->diagnose cycles. Also adds Environment
Layer feature request to docs/annotations.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 19:34:58 +00:00
Bill
850bb8f927 Step 256: batch query endpoint for multi-query round-trips
batchQuery accepts array of sub-queries, returns all results in one
response. Errors in individual sub-queries don't affect others.
22 MCP tools total. 12/12 tests pass.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 16:06:54 +00:00
Bill
92e2f2493a Step 255: symbol-only mode for scope queries (lean vs detailed)
getInScopeSymbols, getCallHierarchy, getDependencyGraph now default
to lean "symbols" mode returning names/IDs only. Pass detailed:true
for full node JSON. 12/12 tests pass.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 16:00:56 +00:00
Bill
ba9da3fd55 Step 254: response budget system (truncation, continuation, pagination)
Optional budget parameter on getAST and getDiagnostics. Responses exceeding
the budget are truncated with continuation tokens for paginated follow-up.
Binary search finds max array elements fitting the budget. Diagnostics sorted
by severity (errors first) before truncation.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 15:54:01 +00:00
Bill
d6941da793 Step 253: diagnostic pipeline integration tests (Phase 9b complete)
End-to-end tests: valid code → zero diags, annotation errors with nodeIds,
quick fix application clearing diagnostics, delta showing removals/additions,
combined parser+annotation streams, severity filtering. Phase 9b: 44/44 tests
across steps 250–253.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 15:48:06 +00:00
Bill
fcccaeb866 Step 252: diagnostic delta streaming (getDiagnosticsDelta, version tracking)
DiagnosticVersionTracker records snapshots and computes added/removed deltas
between versions. Agents can efficiently check if their fix resolved an error
without re-fetching all diagnostics. 21 MCP tools total.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 15:45:31 +00:00
Bill
fbc7b6bce2 Step 251: quick-fix actions as RPC mutations (getQuickFixes, applyQuickFix)
Concrete mutation objects for each diagnostic fix. getQuickFixes returns
reviewable fixes with categories; applyQuickFix applies the fix and reports
whether the diagnostic cleared. Heuristic fixes for missing return and
unused variables. 20 MCP tools total.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 15:41:56 +00:00
Bill
14c2fc218a Step 250: structured diagnostic format (error codes, nodeIds, fix suggestions)
Unified diagnostic stream merging parse errors (E01xx), annotation validation
(E02xx), and strategy violations (E03xx) with severity levels, AST node
references, and machine-applicable fix mutations. New getDiagnostics RPC
method with severity/source filtering. 18 MCP tools total.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 15:36:24 +00:00
Bill
f858925911 Step 249: MCP server integration tests (Phase 9a complete)
End-to-end tests exercising the full MCP stack through MCPBridge:
handshake, 17 tools with valid schemas, AST queries, pipeline execution,
diagnostics, prompts, file CRUD cycle, and compact AST comparison (3% size).
Phase 9a: 64/64 tests pass across steps 245–249.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 15:14:21 +00:00
Bill
fbff0cedd0 Step 248: compact AST response format (subtree, diff, version tracking)
Adds token-efficient AST queries: compact mode (<30% of full size),
subtree extraction by nodeId, version-tracked AST diff after mutations,
and tokenEstimate on all responses. 12/12 tests pass, 17 tools total.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 06:29:17 +00:00
Bill
1131fdf73d Step 247: file operation tools for MCP (read, write, create, diff, list)
Adds 5 workspace-aware file I/O tools to the MCP server so agents can
manage files without a GUI. All paths are resolved against --workspace
root with escape-rejection security. 12/12 tests pass, tool count now 15.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 05:59:02 +00:00
Bill
06706aeacf Step 246: mcp_main.cpp — standalone MCP server entry point
Wires HeadlessEditorState + MCPBridge into a standalone whetstone_mcp
binary that any MCP client (Claude Code, Cursor, etc.) can launch over
stdio. No SDL, no ImGui, no OpenGL — headless only. 12/12 tests pass.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 05:40:32 +00:00
Bill
9fcb5a0c8c Step 245: HeadlessEditorState — agent API surface without ImGui/SDL
Adds the headless agent architecture for Sprint 9 Phase 9a:
- ASTUtils.h: pure AST utilities extracted from EditorUtils.h
- HeadlessEditorState.h: GUI-free state with buffer management
- HeadlessAgentRPCHandler.h: full RPC dispatch (20+ methods)
- step245_test.cpp: 20 tests all passing

Also fixes test compilation errors (NotificationSystem API changes,
AgentRole permissions, DependencyPanel missing include) and adds
SDL2 system library detection fix in CMakeLists.txt.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-11 01:34:27 +00:00
Bill
f63e6a5fdd Sprint 8: add step244 integration test and fix Ubuntu test build 2026-02-10 22:17:12 +00:00
Bill
bfff75397f 3d finished 2026-02-10 13:42:39 -07:00
Bill
686de5a4ef Step 241: text-first defaults and trim headers 2026-02-10 11:22:44 -07:00
Bill
f074f51e7f Add language and plugin enhancement requests 2026-02-10 10:00:27 -07:00
Bill
e62bb3fd8a Add Julia packaging feature request 2026-02-10 09:59:38 -07:00
Bill
c09a76d74d Add Julia feature requests and roadmap update 2026-02-10 09:56:03 -07:00
Bill
9b9244e0f8 Add language support roadmap and feature requests 2026-02-10 09:51:23 -07:00
Bill
5670ab7f5e Step 234: Add session pipeline tests 2026-02-10 09:09:38 -07:00
Bill
423db8cb22 Step 233: Add training data pipeline 2026-02-10 08:47:07 -07:00
Bill
b68d24c881 Rename session trace header 2026-02-10 08:43:19 -07:00
Bill
2491ece973 Step 232: Add session trace converter 2026-02-10 08:41:59 -07:00
Bill
d909b07301 Step 231: Add session anonymizer 2026-02-10 08:36:22 -07:00
Bill
835b90bbac Step 230: Enhance session recorder 2026-02-10 08:33:19 -07:00
Bill
718c99fef9 Step 229: Add tool definition tests 2026-02-10 08:23:48 -07:00
Bill
7e329b6a2e Step 228: Add prompt templates 2026-02-10 08:20:59 -07:00
Bill
1d91c49b15 Step 227: Open-source model tool definitions 2026-02-10 08:19:07 -07:00
Bill
8a15b27380 Step 226: OpenAI tool definitions 2026-02-10 08:14:19 -07:00
Bill
23117eb97e Step 225: Claude tool definitions 2026-02-10 08:12:16 -07:00
Bill
82bc20b5c8 Step 224: evaluation harness tests 2026-02-10 08:06:04 -07:00
Bill
3e02f004f0 Step 223: evaluation runner CLI 2026-02-10 08:04:00 -07:00
Bill
1ce68176d1 Step 222: workflow evaluation tasks 2026-02-10 08:02:24 -07:00