cflmy and Cursor
1730f26fc7
release: v2.2.0 call one function in any .mq.md
...
The same page is the document, the knowledge and the program. marqdo call runs that function with arguments, so reuse, tools and tests do not need a second skill system.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-30 09:07:23 +08:00
cflmy and Cursor
303c06f773
feat(workbench): freeze agent protocols and ship ResourceView v1
...
Lock M-A0 Error/Tool/Checkpoint/Output/Context fixtures with capability bits off, and land M-A3 static md/txt/code ResourceView (no-exec) behind resource.view.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-27 10:59:59 +08:00
cflmy and Cursor
03396afd02
feat(workbench): land M00–M06 runtime contracts and read-only view
...
Add capabilities/selection/usage/DocumentView/locator/RunContext with
honest probes, LLM messages+cancel, sse-live cancel bridge, and CI
workbench_contracts gate; record design evidence for verified slices.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-25 10:55:08 +08:00
cflmy and Cursor
dc853fb580
release: bump to 2.1.2 (EK production chain + View V0–V3 bar)
...
Ship evidence/closure invalidation, verify/adjudicate, reuse_execute and
agent bind/publish gates, document.extract, KnowledgePresentation + SSE
page_token, and upload max_bytes with the 45-test offline gate.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-24 22:52:03 +08:00
cflmy and Cursor
8a6d09d582
feat: pure extract, bookstore story subset, and project scan caps
...
Add artifact.extract_pure for restricted expression lifting, a runnable
bookstore tax story fixture, and G12 discovery limits so oversized or deep
trees fail with explicit diagnostics instead of hanging.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-23 23:27:05 +08:00
cflmy and Cursor
dcd3d329d4
feat: land S1–S6 executable-knowledge slices (NarrativeRef through CI)
...
Add comment-line NarrativeRef steps, generation CURRENT snapshots, frontmatter
sources/applicability, digest-guarded artifact.patch, lock change classes, and
an offline CI gate so the documented S0–S6 path has reviewable evidence.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-23 23:24:18 +08:00
cflmy and Cursor
ecbdd18303
feat: executable-knowledge S0–S1 gate plus effect and MLSP reuse subset
...
Establish author baseline fixtures and offline tests, make SemanticProject
the sole authority for graph edges, classify named steps, and tighten
local reuse so unknown effects are not pure and MLSP binds by SymbolId.
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-10-23 22:52:06 +08:00
cflmy
eac9b45263
feat: compile executable knowledge semantic artifacts
2026-10-23 09:16:52 +00:00
cflmy
6bba0be6d2
feat(mlsp): M6 T1 会话内陈旧检测 — repair_apply 前后对照 + session.summary
...
机制:repair_apply 内同一作用域已同时持有 source(改前) 与 new_source(改后),
所以「描述是否还成立」可纯内存判定,无需 git/台账/exec:
stale ⟺ 标签文本未变 ∧ payload 指纹变了
按 (标签文本, 出现序号) 配对 → 移动代码不误报。
协议:
请求加可选 check_labels(默认 false,零开销零噪音)
响应加 labels.stale[{label,line,code_past,code_now}] + newly_seen
新增 session.summary → {labels_stale_seen, labels_fixed, labels_outstanding, items},
返回后清零(客户端收尾调用一次即闭合,不主动打屏)
取舍:
- 标签变了 → 不报(作者改描述即"已复核"信号)
- 新标签 → 不报(无历史可比)
- 指纹用 Debug:只需进程内一致;跨版本稳定指纹留给 T2/T3
- code_past/code_now 用 view::expr_display 渲染成人可读的表面语法
实测:
改 payload 0.1→0.2、标签不变 → stale 且带 code_past="100 * 0.1" / code_now="100 * 0.2"
同时改标签 → stale: []
session.summary → stale_seen=1, outstanding=1, items=["按税率算总价"]
测试:3 个集成(含反向验证"改描述即复核不误报");
全量 cargo test --release:590 passed / 0 failed(M10 后 587 + 3)
文档:doc/design/change-m6-t1-session.md
2026-10-22 15:19:17 +08:00
cflmy
fd0cfcb566
feat(labels): M9 两级词表检索 — labels CLI + MLSP labels.search
...
第一级(内置,数据常量):六组关键词 —— 未完成/假数据/示例/实验/废弃/待校验,
pending = 未完成 ∪ 假数据。作用是默认查询、LLM 范例、跨项目可比。
第二级(参数):--grep 替换,--also-builtin 并集,--group 选组。
**词表是数据不是语法**:扩充它不改任何解析规则。
CLI:
marqdo labels . # pending 口径
marqdo labels . --grep "联调" # 自定义替换
marqdo labels . --grep "联调" --also-builtin # 并集
marqdo labels . --group 示例,实验
marqdo labels . --count / --json
MLSP:{"method":"labels.search","params":{path|source, grep?, also_builtin?, group?}}
→ {"ok":true,"words":[…],"count":N,"hits":[{line,unit,label,words,has_payload}]}
与 CLI 共用同一份 label::search。
检索语义 v1:子串匹配(CJK 无词边界)、多词 OR、英文大小写不敏感;
误判只影响计数,输出带描述原文可人工判断;不做否定语法(留 v1.1)。
测试:4 个集成(pending=2 / grep 替换=1 / also-builtin 并集=3 / MLSP 返回 hits);
全量 cargo test --release:587 passed / 0 failed(M4 后 583 + 4)。
实测:
pending → 2 命中("待开发"、"模拟"两处)
--grep 联调 → 1 命中
--grep 联调 --also-builtin → 3 命中
文档:doc/design/change-m9-label-search.md
2026-10-22 15:16:39 +08:00
cflmy and Cursor
1a729eb9ec
P4 验证执行:(a)锚点消融✅ /(c)契约消融✅ /(d)防漂移✅ /(e)守卫演练✅ ;(b)盲测 deepseek-flash 完成未达标、mimo-v2.6-pro 重跑中
...
- perf-validation.md:五实验协议/判据/材料(2026-09-23 自设计)
- perf-validation-report-2026-09-23.md:汇总报告(判据逐项对照 + 诚实结论)
- tests/blind_experiment.rs + tests/blind/:盲测 harness(查询通道/修复回路/usage 计量)+ 冻结材料(6 任务书 + docs-bundle)
- tests/contract_ablation.rs:四边界×三情形消融(拦截 100%/误报 0%/零回归)
- tests/drift_drill.rs:9 漂移必报 + 2 负对照
- src/mlsp.rs:syntax 卡片自持化(semantics/rule/sample/doc_quote)
- three-problems.md §5.1 判据矩阵逐行判定(问题 1✅ 2✅ 3❌ );three-problems-plan.md Phase 4 落地记录
- 全量干净回归零失败
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-09-23 05:06:30 +08:00
cflmy and Cursor
5b599125d1
feat(repair): T3.3 修复护栏 + golden 20 例 —— 机械护栏(有界行编辑·越界必拒·abstain)+ mlsp repair_apply
...
- src/repair.rs:行编辑有界应用(replace/delete/insert;触碰行必须落在靶点范围内,
越界 ⇒ 整体拒绝 + 越界清单,源零字节不改 = abstain;表尾追加允许 end+1)
- mlsp repair_apply:护栏的 MLSP 出口(ranges + edits → 新源;越界报 mlsp.repair_out_of_scope)
- tests/repair_loop.rs:golden 20 例(10 骨架 × 2)走完整修复循环
(诊断 → 靶点 → 金修复行编辑自动推导 → 范围自检 = 「易锚定」机器验证 → 应用 → 复验)
+ 护栏负例(越界必拒 + 源零字节不改 + 部分越界全拒)
- 待做:模型修复实验(把金修复换成模型编辑,测 ≥80% 修复率)
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-09-22 20:13:41 +08:00
cflmy and Cursor
db5f0e830b
feat(mlsp): Phase 3 MLSP for AI —— locate/syntax/validate/repair_targets/schema(T3.1/T3.2 + T2.6)
...
- src/mlsp.rs: 行分隔 JSON 协议 + 五 handler;语法唯一事实来源 = CORE_CONSTRUCTS(只加检索别名)
- repair_targets 带有界护栏策略(anchored-local / 越界必拒 abstain / max_attempts=2)
- schema 契约导出(T2.6):形参/返回/字段契约 + 升参清单 + 锚点
- marqdo mlsp 子命令:stdio 行分隔 JSON 服务
- tests/mlsp.rs 7 例全绿;全量回归零失败(含 262 gold)
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-09-22 19:44:35 +08:00