From 76ec4624a841133c016463fb95fed59292d2a4fc Mon Sep 17 00:00:00 2001 From: g1nation Date: Tue, 5 May 2026 16:11:05 +0900 Subject: [PATCH] [UX] Mass overhaul: Remove internal logs/questions and focus on actionable file-level guidance --- ...connectai-내-질문에-대한-답변이-잘-정리.md | 16 +++++++++++++++ docs/records/ConnectAI/chronicle.config.json | 4 ++-- docs/records/ConnectAI/timeline.md | 3 +++ package.json | 2 +- src/utils.ts | 20 +++++++++---------- 5 files changed, 31 insertions(+), 14 deletions(-) create mode 100644 docs/records/ConnectAI/bugs/BUG-0004-volumes-data-project-antigravity-connectai-내-질문에-대한-답변이-잘-정리.md diff --git a/docs/records/ConnectAI/bugs/BUG-0004-volumes-data-project-antigravity-connectai-내-질문에-대한-답변이-잘-정리.md b/docs/records/ConnectAI/bugs/BUG-0004-volumes-data-project-antigravity-connectai-내-질문에-대한-답변이-잘-정리.md new file mode 100644 index 0000000..43429c9 --- /dev/null +++ b/docs/records/ConnectAI/bugs/BUG-0004-volumes-data-project-antigravity-connectai-내-질문에-대한-답변이-잘-정리.md @@ -0,0 +1,16 @@ +# Bug: /Volumes/Data/project/Antigravity/ConnectAI 내 질문에 대한 답변이 잘 정리되서 알려주긴 하는데 focused... + +## Date +2026-05-05 + +## Symptom +/Volumes/Data/project/Antigravity/ConnectAI 내 질문에 대한 답변이 잘 정리되서 알려주긴 하는데 focused 된 답변이 아니라 살짝 다른 이야기도 섞여 있는 것 같은 느낌이 들어. 어디가 문제인지 확인해줘. + +## Cause +Captured automatically from the current conversation. Confirm root cause during follow-up review if needed. + +## Fix +저는 질문자님의 피드백을 주의 깊게 들었습니다. 답변이 집중되지 않고 다른 이야기가 섞여 나온다는 것은 정보의 추출 과정에서 맥락이 충분히 정제되지 않고 노이즈가 유입되고 있다는 신호로 해석됩니다. ## 간단 요약 이는 주로 정보 검색 결과의 정밀도 부족이나, 여러 출처에서 나온 정보를 통합할 때 경계 설정이 느슨하게 적용되어 발생하는 현상으로 보입니다. 현재 시스템은 여러 지식 소스를 통합하는 과정에서 세부적인 내용의 경계를 명확히 구분하지 못하고 있거나, 검색된 정보의 밀도 차이를 충분히 고려하지 못해 발생하는 문제입니다. ## 상세 답변 이 문제는 지식 검색 및 통합 단계에서 발생할 가능성이 높습니다. 특히, 여러 문서나 지점에서 정보를 가져올 때 각 정보의 중요도와 맥락적 연관성을 충분히 차등 부여하지 못하면, 핵심 내용 외의 주변 정보가 함께 추출되어 답변에 포함될 수 있습니다. 이는 의도된 '고밀도' 정보 추출 과정이 기대만큼 정교하게 작동하지 않았거나, 혹은 검색 결과의... + +## Prevention +Keep automatic records tied to the active project and verify the relevant test or reproduction path. diff --git a/docs/records/ConnectAI/chronicle.config.json b/docs/records/ConnectAI/chronicle.config.json index 19eee6d..cbd6bbf 100644 --- a/docs/records/ConnectAI/chronicle.config.json +++ b/docs/records/ConnectAI/chronicle.config.json @@ -6,6 +6,6 @@ "description": "Auto-detected from the local project path in the conversation.", "corePurpose": "Capture project direction, architecture discussion, decisions, and development notes as Markdown.", "detailLevel": "standard", - "createdAt": "2026-05-05T06:33:55.096Z", - "updatedAt": "2026-05-05T06:33:55.103Z" + "createdAt": "2026-05-05T06:57:16.213Z", + "updatedAt": "2026-05-05T06:57:16.217Z" } diff --git a/docs/records/ConnectAI/timeline.md b/docs/records/ConnectAI/timeline.md index 7ff6a38..03d019b 100644 --- a/docs/records/ConnectAI/timeline.md +++ b/docs/records/ConnectAI/timeline.md @@ -48,3 +48,6 @@ ## 2026-05-05 - Auto bug record created: bugs/BUG-0003-volumes-data-project-antigravity-connectai-내-질문에-대한-답변이-잘-정리.md + +## 2026-05-05 +- Auto bug record created: bugs/BUG-0004-volumes-data-project-antigravity-connectai-내-질문에-대한-답변이-잘-정리.md diff --git a/package.json b/package.json index 241f091..66ec3ee 100644 --- a/package.json +++ b/package.json @@ -2,7 +2,7 @@ "name": "astra", "displayName": "Astra", "description": "The personal intelligence layer for Antigravity and VS Code. A private cognitive partner for deep project context, memory, and proactive strategic decision-making.", - "version": "2.76.3", + "version": "2.76.4", "publisher": "g1nation", "license": "MIT", "icon": "assets/icon.png", diff --git a/src/utils.ts b/src/utils.ts index c470989..4e38be1 100644 --- a/src/utils.ts +++ b/src/utils.ts @@ -152,18 +152,16 @@ Core behavior: - Do not say "upload the source code", "a folder path is not enough", or "please provide files" before attempting , , or a safe listing command for the provided path. - If the path cannot be accessed after trying, explain the access failure and only then ask for an upload or workspace connection. - After action results are available, summarize the actual findings directly. -- [STRICT NEGATIVE CONSTRAINT] Do NOT output internal reasoning headers or process labels such as: "간단 요약", "요청 요약", "사용자 의도 추론", "프로젝트 기록 대상 확인", "핵심 확인 질문", "Astra 판단", "내가 보는 위험", "다음 한수", or repetitive "근거" blocks. These are for your internal logic only and must NEVER be visible to the user. -- For all complex analyses, use ONLY these three visible markdown headings: ## Verdict (결론), ## Proof (근거), ## Actionable Plan (실행 가이드). -- 1. ## Verdict (결론): Place the final judgment at the very top. 2-5 simple sentences. No fluff. -- 2. ## Proof (근거): Provide supporting facts and objective risk criteria (Acceptable Loss vs. Critical Stop). Do not repeat this section. -- 3. ## Actionable Plan (실행 가이드): Concrete next steps with "If X, then Y" logic. -- Do not append any summary, conclusion, or meta-commentary at the end of the response. Stop immediately after the Actionable Plan. -- Use paragraph-style formatting for Verdict and Proof. Use a numbered list for Actionable Plan only when sequence matters. -- For simple replies or quick updates, do not use any headings. Just answer conversationally. -- Avoid inflated consulting language. Be a direct engineering partner. Give an opinionated verdict, then explain tradeoffs and the next small decision. -- If the user sounds unsure, reassure them briefly, then return to concrete diagnosis. Do not imply the issue is the user's intelligence. +- [CRITICAL: NO INTERNAL LOGS] Never output
blocks, "Second Brain Trace", or "Debug JSON". These are strictly for internal debugging and must NEVER be visible to the user. +- [STRICT NEGATIVE CONSTRAINT] Prohibit all "interview-style" confirmation questions and redundant sections: "간단 요약", "요청 요약", "사용자 의도 추론", "프로젝트 기록 대상 확인", "핵심 확인 질문", "Astra 판단", "내가 보는 위험", "다음 한수", "근거". +- For all responses, provide a direct answer immediately without labels if possible, or use only these three headers for complex reports: ## Verdict (결론), ## Proof (근거), ## Actionable Plan (실행 가이드). +- 1. ## Verdict (결론): Direct, opinionated answer. If the user asks for a file or logic location, name it here immediately. +- 2. ## Proof (근거): Brief supporting evidence or reasoning. Keep it minimal. +- 3. ## Actionable Plan (실행 가이드): Specific To-Do items. List exact file paths to edit or commands to run. If there's no next action, omit this section. +- Stop immediately after the final section. No meta-commentary like "I hope this helps". +- Be an engineering partner, not a consultant. Use technical precision over polite filler. - [STRICT RULE: NO EMOJIS] Professional text-only output. -- [STRICT RULE: UNIQUE HEADINGS] Each markdown heading must be unique and appear exactly once. +- [STRICT RULE: UNIQUE HEADINGS] Each markdown heading must appear exactly once. - [STRICT RULE: NO EMOJIS] Do not use any emojis, icons, or pictorial symbols in your response. Keep the tone professional and text-based only. - [STRICT RULE: UNIQUE HEADINGS] Do not repeat section titles. Ensure each markdown heading is unique and serves a specific structural purpose. - Do not use grand labels like "final execution mandate", "engineering standard", "knowledge distiller", or "Antigravity's yardstick" unless the user explicitly asks for that style.