v0.1.0-rc1

Changelog

• 9a88ec7 Merge pull request #100 from Aruing/release/sync-production-2
• a06bcf0 ci: GoReleaser + release workflow——production tag 驱动发版（0.1.0-beta22 步骤 4） (#99)
• c4ceae6 Merge pull request #97 from Aruing/release/sync-production-1
• 5b9fdf4 fix(ci): macOS 冒烟 SIGPIPE——version 输出先落变量再匹配（#97 全链验证抓到） (#98)
• 3107b66 ci: release-check workflow——production PR 发布检查（0.1.0-beta22 步骤 3） (#96)
• 769a46d chore: 模块路径替换 aruing → github.com/Aruing/Aruing（0.1.0-beta22 步骤 2） (#95)
• 2883cf6 feat(cli): 版本注入改造——ldflags 三变量注入 + make version 构建后跑（0.1.0-beta22 步骤 1） (#94)
• e213886 docs(state): close 0.1.0-beta21, point next at 0.1.0-beta22（发布管线） (#93)
• 1ad37a8 feat(tui): 留白统一与称呼开关——spacing 归一 + labels（0.1.0-beta21） (#92)
• 1582512 chore: .gitignore 加 .DS_Store（macOS 系统噪音不入库）
• 2a14cca docs: README 补示例截图 + 新增 CONTRIBUTING (#91)
• 46dc11d Merge pull request #90 from Aruing/docs/skill-discipline
• d0c49da docs(pr-description): label 落上核对为硬规则——创建命令可能静默丢 label
• 3ade043 docs: 吸收 dsh skill 纪律——取代审计 / 文档时态嗅探 / README 双语硬规则
• f2ffec8 docs: 开源文档专业化——README 重写 stage/TUI/Requirements/badges，入口表补全 (#89)
• c00a479 chore(config): 删未用的 env-only Load（审计裁决） (#88)
• 735adc2 docs: beta18/20 关闭后收尾补全——README skills 入口过时 + milestone-close 合并确认规则 (#87)
• 6577964 docs(state): close 0.1.0-beta18 + 0.1.0-beta20, reopen candidate window (#86)
• cc22b2e docs(skills): 反思机制 skill——卫生扫描 + 守规审计（beta18 第 5+6 项落地） (#85)
• 427af95 feat(tui): 主题 YAML 完整化——全样式项覆盖（0.1.0-beta20） (#82)
• 65f6dc4 docs(skills): 真集群测试纪律 skill + investigate 挂起强测 case（beta18 step2） (#84)
• d3a7a7a ci(scenarios): cases/ 同集群多 case 协议 + smoke-all 改造（beta18 step1） (#83)
• 973b074 ci(scripts): smoke-all 全场景自检 + self-check skill（beta18 第 4 项落地） (#81)
• dd6668b docs(state): close 0.1.0-beta19, reopen candidate window
• 53d2eaf feat(agent): 调查阶段挂起——investigate clarify（0.1.0-beta19） (#80)
• 99c0235 docs(skills): 里程碑收尾 skill + PR 元数据自动化（beta18 1/2 落地） (#79)
• 5b1384b docs(state): close 0.1.0-beta17, reopen candidate window (no 0.1.0 close evaluation per maintainer)
• c210730 feat(tools): logs 时间游标——evidence.read 时间窗切片（beta17 step1） (#78)
• 97507b0 docs(state): close 0.1.0-beta16, reopen candidate window
• fcc06e8 feat(tools): 非表格输出行级分页（beta16 step1） (#77)
• d98f550 docs(state): close 0.1.0-beta15, reopen candidate window
• 2f03b1c fix(tui): app mode renders to real terminal with alt-screen
• 9f67b69 fix(tui): app mode tty precheck + explicit stdin binding
• c9e7102 chore(make): UI variable for chat mode passthrough
• 820e2d8 feat(cli): interactive mode switch via --ui / tui.mode (0.1.0-beta15 step 3) (#76)
• 645f8ed docs: restore project-state emptied by #74 (#75)
• aa86d71 feat(tui): inline round divider + glamour markdown (0.1.0-beta15 step 2) (#74)
• 1872142 feat(tui): inline chat engine (pi scrollback style) (0.1.0-beta15 step 1) (#73)
• 9b59122 feat(tui): glamour markdown + theme config (0.1.0-beta14 step 3) (#72)
• 41cfb6e feat(tui): bubbletea inline chat skeleton (0.1.0-beta14 step 2) (#71)
• e56e54d feat(cli): kubectl/context visibility in startup banner (0.1.0-beta14 step 1) (#70)
• f263167 docs: archive 0.1.0-beta13 (evidence navigation L3 leg B) (#69)
• 379242c feat(tools): evidence navigation via evidence.read + round-scoped index (0.1.0-beta13 step 2) (#68)
• f435ef6 refactor(tools): extract table summary into internal/tools/summary (0.1.0-beta13 step 1) (#67)
• cd3d32b feat(tools/k8s): large-table PCA anomaly sampling (0.1.0-beta12 step 1) (#66)
• f3d26f2 docs(state): prioritize L3 (tool-output drill/paging) as diagnosis-accuracy blocker
• 24b5673 docs(state): close 0.1.0-beta11 (merged #65, plan archived)
• c12cfe3 feat(tools/k8s): structured tool output projection (0.1.0-beta11 step 1) (#65)
• 7be0f84 docs: arc
• 881385c docs(state): close 0.1.0-beta10 (merged #64, plan archived)
• 4b1861b feat(scenarios): kind reproducible scenario harness (0.1.0-beta10) (#64)
• f161b10 docs: strict sync architecture/project-state with merged beta9 code
• 6de960d feat(resolve): clarify suspension + resume (0.1.0-beta9 step 1) (#63)
• 1470a4f docs: current documents with config
• 930664c docs: update readme
• 1513dd1 docs: current project state
• bcbcd13 fix(cli): print config path and model banner; ignore playground/ (#62)
• 1b97d6a feat(config): YAML config + env overlay; pure-Chinese Go comments (#61)
• 7ab2b3e docs: close 0.1.0-beta7 after rehydrate smoke (beta7-2) (#60)
• ce24430 feat(agent): post-compaction range rehydrate (beta7-1) (#59)
• c336dd8 fix(llm): retry empty responses; Tower recover + debug progress (#58)
• d6f7a40 docs: close 0.1.0-beta6 after smoke (beta6-3) (#57)
• 9f37e61 feat(agent): inject prior_run_details from RunLedger (beta6-2) (#56)
• c3e7d4e feat(session): process-local RunLedger for escalate Report+Evidence (#55)
• b88362f Update README.md
• d48a5f3 docs: close beta5-fix-2 after smoke (T-shallow-4) (#54)
• 83d5d5b fix(agent): widen baseline tool budget; auto-escalate on cap; evidence discipline (#53)
• 9a48efe fix(agent): inject cluster_resources into Tower baseline via light recon (#50)
• 5d3e26e docs: close beta5-fix-1 and record baseline escalate known gap (#49)
• 6f1d9c6 docs: sync project-state after beta5-fix-1 T-obs-1/2/4 merge (#48)
• 8c6702a fix(agent): share Resolver evidence raw budget without 2000-char wall (#47)
• 26fb3b5 fix(agent): share Tower observation raw budget across tools (#46)
• d3bfd52 fix(agent): feed Evidence.Raw into Tower baseline observations (#45)
• 4eafbd2 docs: close beta5 milestone and refresh project state (#44)
• 0e12a3c feat(agent): L2 handoff compact and ModeCheckpoint (beta5-5 PR-B) (#43)
• a75e540 feat(agent): tower explain-prior context budget (beta5-5 PR-A) (#42)
• b643ffd docs: sync README and project-state after beta5-4 chat (#41)
• 451bd08 feat(cli): add aruing chat via Session.Turn + Tower (beta5-4) (#40)
• 3a06e07 feat(agent): add Tower baseline call_tool loop (beta5-3) (#39)
• aa7678c docs: complete production struct field comments (#38)
• f3f652e feat(agent): minimal Tower reply/escalate (beta5-2) (#37)
• 4a5d275 feat(session): add Session/Message/Turn skeleton (beta5-1) (#36)
• b52c77e docs: close beta4 milestone and set current to beta5 (#35)
• 1bd1332 chore(ci): switch pr-agent to DeepSeek official API (v4-pro) (#34)
• aaedecd docs: sync post-beta4 and drop redundant PR list from project-state (#33)
• 00ffcdf feat(tools): add DiagnosticPolicy for diagnostic exec (beta4 全景-5) (#32)
• c73b4b9 feat(agent): strengthen Planner reflection prompt (beta4 全景-4) (#31)
• 392ffbb feat(agent): discover cluster resources and feed Planner (beta4 全景-3) (#30)
• bd13531 feat(agent): reuse resolve-phase evidence in investigation (beta4 全景-2) (#29)
• 2f0ffcf feat(agent): thread Query into Verifier (beta4 诊断全景-1) (#28)
• e36f0e6 feat(cli): render evidence detail chain in Markdown report (beta3 多轮-4b) (#27)
• 63b5314 feat(agent): tolerate tool failures during investigation (beta3 多轮-4a) (#26)
• bb44f9c feat(cli): show actual kubectl command in progress trace (#25)
• fd044f6 fix(agent): continue investigation when all hypotheses refuted (#24)
• 4db88dc feat(agent): adaptive Planner prompt and raise investigate budget (beta3 多轮-3) (#23)
• e6b5bbf feat(agent): orchestrated investigation loop (beta3 多轮-2) (#22)
• 28392ea feat(agent): thread PlanState into Planner interface (beta3 多轮-1) (#21)
• ff6ef45 docs: mark 0.0.1-beta2 complete after end-to-end validation (#20)
• 3deff24 feat(cli): render Markdown diagnosis report by default (#19)
• 7462ec3 feat(config): centralize env loading and minimal LLM error hints (#8) (#18)
• 808d1f5 feat(agent): replace FakeReporter with LLMReporter (#7) (#17)
• 6f0e977 feat(agent): replace FakePlanner and FakeVerifier with LLM implementations (#16)
• 73d015b feat(agent): orchestrated resolve loop and LLMResolver (#4b) (#15)
• f6209e1 feat(tools): add readonly Policy and optional k8s wiring (#4a) (#13)
• 89603ce docs: record linear Orchestrator as temporary single-turn driver (#12)
• 19b449a feat: add shell-less Kubernetes tool and discoverable specs (#11)
• fcf8421 fix: correct tool readonly constraint - interface does not limit read/write (#10)
• 3775e7e docs: add aruing-pr-description skill (#9)
• 19b8973 docs: add repo documentation per aruing-docs skill (#8)
• a9b74a8 docs: add aruing-docs skill for repo documentation conventions (#7)
• 8fad9ea fix(parser): validate ref uniqueness and retry on inconsistent LLM output (#6)
• dc09495 feat: replace FakeParser with LLMParser (#5)
• 39abbd4 ci: add pr-agent auto review (#4)
• 924a16e feat: llm client (#3)
• ddc2181 Feat/check action (#2)
• 0db3650 Feat/init mvp (#1)
• bbaddbe Initial commit