docs(spec): Sprint 1 plan — 4 SPECs audit-ready#118
Merged
Conversation
…-TELEGRAM + TOOLS-WEB + SCHEDULER) GOOSE Sprint 1 (Hermes Agent 추월 8주 로드맵의 첫 sprint) plan phase 완료. 4 SPEC 모두 audit-ready 상태 도달 (총 79 EARS REQs + 60 ACs). ## 신규 SPEC 4종 - SPEC-GOOSE-LLM-ROUTING-V2-001 (P1, 14 REQ + 11 AC) - 15 direct adapter cost/latency/capability-aware 라우팅 + 수동 fallback chain - OpenRouter 미도입 결정 (6개월 후 재평가) - self-audit CONDITIONAL_GO (REQ-003/006 minor gap, 사용자 수용 후 audit-ready) - SPEC-GOOSE-MSG-TELEGRAM-001 (P0, 25 REQ + 11 AC) - Telegram bot ingress + send tool, BRIDGE-001 위 wiring - plan-auditor iter-1 CONDITIONAL_GO → 5 clerical fix → iter-2 PASS (Score 0.91) - SPEC-GOOSE-TOOLS-WEB-001 (P0, 18 REQ + 18 AC) - 8 web 도구 (search/browse/RSS/Wikipedia/ArXiv/maps/wayback/HTTP) - self-audit PASS - SPEC-GOOSE-SCHEDULER-001 (P0, 22 REQ + 20 AC) - Proactive Ritual Scheduler (cron + timezone + holiday + pattern learning hook) - plan-auditor iter-1 PASS (Score 0.92) - RITUAL-MORNING/EVENING (Sprint 3-4) 인프라 ## 산출물 - 20 SPEC 문서 (spec/plan/acceptance/spec-compact/research × 4) - 3 plan-audit reports (.moai/reports/plan-audit/) - 2 status.txt (MSG-TELEGRAM 신규, SCHEDULER 갱신) ## 전략 결정 - LLM 어댑터: 현 15 direct 유지 (OpenRouter 미도입) - 도구 카탈로그: 75개 (멀티미디어/사주/한국특화 별도 백로그, Hermes 40+ 대비 1.9배) - 메시징 우선순위: Telegram → (Sprint 2: Email) → Discord/Slack - 자동 생성: 3-tier (skill + agent + team), Sprint 2-4에서 단계적 도입 ## 다음 단계 각 SPEC /moai run 진입 가능. 권장 순서: 1. TOOLS-WEB-001 (웹 도구 인프라) 2. MSG-TELEGRAM-001 (사용자 채널) 3. LLM-ROUTING-V2-001 (라우팅 v2) 4. SCHEDULER-001 (proactive 인프라) 🗿 MoAI <email@mo.ai.kr>
GoosLab
added
type/docs
|
Warning Rate limit exceeded
To keep reviews running without waiting, you can enable usage-based add-on for your organization. This allows additional reviews beyond the hourly cap. Account admins can enable it under billing. ⌛ How to resolve this issue?After the wait time has elapsed, a review can be triggered using the We recommend that you space out your commits to avoid hitting the rate limit. 🚦 How do rate limits work?CodeRabbit enforces hourly rate limits for each developer per organization. Our paid plans have higher rate limits than the trial, open-source and free plans. In all cases, we re-allow further reviews after a brief timeout. Please see our FAQ for further information. ℹ️ Review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro Run ID: 📒 Files selected for processing (4)
Walkthrough이 PR은 GOOSE 에이전트의 세 가지 주요 기능(LLM 라우팅 v2, Telegram 메시징, 웹 도구)과 스케줄러 구현을 위한 specification 문서들을 추가합니다. Telegram과 스케줄러 spec에 대한 감사 리뷰 보고서도 포함되어 있습니다. ChangesSPEC-GOOSE-MSG-TELEGRAM-001 (Telegram 메시징 통합)
SPEC-GOOSE-LLM-ROUTING-V2-001 (라우팅 v2 decorator)
SPEC-GOOSE-SCHEDULER-001 (Daily Companion 스케줄러)
SPEC-GOOSE-TOOLS-WEB-001 (웹 도구 카탈로그)
Estimated Code Review Effort🎯 4 (Complex) | ⏱️ ~50 minutes 이유: 이 PR은 18개 이상의 specification 문서를 추가하며, 각 spec은 다음을 포함합니다:
reviewer는 다음을 확인해야 합니다:
문서의 양이 크고 상호 참조가 많아 종합적인 검토가 필요합니다. Suggested Labels
🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 9
🧹 Nitpick comments (9)
.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/plan.md (1)
86-102: ⚡ Quick winASCII/의사코드 fenced block에 language를 지정해 주세요.
Line 86, Line 112 코드 블록이 language 없이 열려 markdownlint(MD040) 경고가 납니다.
text를 지정하면 린트/렌더링이 안정적입니다.제안 diff
-``` +```text ┌────────────────────────────────────────────────────────┐ ... └────────────────────────────────────────────────────────┘ -``` +``` -``` +```text Start(ctx, deps): 1. config.Load() → yaml + keyring token ... 9. block until ctx.Done() -``` +```Also applies to: 112-124
🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/plan.md around lines 86 - 102, Two fenced code blocks (the ASCII diagram block near "cmd/goose/cmd/telegram.go (cobra: setup/start/...)" and the pseudocode block beginning "Start(ctx, deps):") are missing a language tag and trigger markdownlint MD040; update their opening fences from ``` to ```text so the blocks are explicitly marked as plain text (apply the same change for the second block around the Start(ctx, deps) pseudocode)..moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/spec-compact.md (1)
29-46: ⚡ Quick winFenced code block에 language 지정이 필요합니다.
Line 29 코드 블록에 language가 없어 markdownlint(MD040) 경고가 발생합니다. 문서 렌더링/하이라이팅 일관성을 위해
text또는bash등 명시를 권장합니다.제안 diff
-``` +```text internal/messaging/telegram/ ├── bootstrap.go # Start(ctx, deps) entry — daemon hook ... internal/daemon/bootstrap.go # MODIFY: messaging.Start 호출 추가 -``` +```🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/spec-compact.md around lines 29 - 46, The fenced code block containing the directory listing (the block starting with "internal/messaging/telegram/") lacks a language tag and triggers markdownlint MD040; update that block by adding a language identifier such as text (i.e., change the opening fence from ``` to ```text) so the snippet is consistently rendered/highlighted in the spec file (SPEC-GOOSE-MSG-TELEGRAM-001/spec-compact.md)..moai/specs/SPEC-GOOSE-TOOLS-WEB-001/research.md (1)
244-247: ⚡ Quick winSSRF 방지 로직 명확화 필요.
Line 246에서 "본 SPEC 은 이런 IP 를 hard block 하지 않음" 이라고 했다가, line 247에서 "default blocklist 에 포함하여 explicit allow 필요" 라고 하여 약간 모순적입니다.
의도가 "hard block은 아니지만 default blocklist에 포함되어 있어서 사용자가 명시적으로 허용해야 호출 가능" 이라면, 이를 더 명확하게 표현하는 것이 좋겠습니다.
📝 명확성 개선 제안
-본 SPEC 은 이런 IP 를 hard block 하지 않음 (사용자가 의도적으로 로컬 서비스 호출 가능). 그러나 **default blocklist 에 포함하여 explicit allow 필요** 하도록 설계. +본 SPEC 은 이런 IP 를 시스템 레벨에서 hard block 하지 않으나, **default blocklist 에 포함**하여 사용자가 명시적으로 허용(blocklist 에서 제거 또는 override)해야만 호출 가능하도록 설계.🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/research.md around lines 244 - 247, Clarify the apparent contradiction in section "9.6 SSRF 방지": update the text around "본 SPEC 은 이런 IP 를 hard block 하지 않음" and the following sentence about "**default blocklist 에 포함하여 explicit allow 필요**" to state clearly that http_fetch will not be universally hard-blocked but that these IPs are placed in a default blocklist (so calls are blocked by default) and require an explicit allow action via the PERMISSION-001 Confirmer (include how the Confirmer indicates scope), and mention examples like 127.0.0.1, localhost, 169.254.169.254 and private CIDRs to make the intended behavior unambiguous..moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec.md (3)
347-355: 💤 Low valueDecision tree 블록에 language identifier 추가 권장
의사결정 트리가 fenced code block으로 작성되었지만 언어 식별자가 누락되었습니다.
text식별자를 추가하면 렌더링이 개선됩니다.제안된 수정
-``` +```text 1. v1.Route(req) → primary/cheap 결정 2. policy.Mode == PreferQuality && chain 없음 → v1 결정 그대로 (v1:simple|complex)🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec.md around lines 347 - 355, The fenced decision-tree code block that starts with "1. v1.Route(req) → primary/cheap 결정" is missing a language identifier; update the opening triple-backtick to include the text language identifier (```text) so the block renders correctly (affects the decision tree lines referencing v1.Route(req) and policy.Mode conditions).
319-330: 💤 Low valuePackage structure 블록에 language identifier 추가 권장
패키지 구조 diagram이 fenced code block으로 작성되었지만 언어 식별자가 누락되었습니다. 디렉터리 구조이므로
text식별자를 추가하는 것이 적절합니다.제안된 수정
-``` +```text internal/llm/router/v2/ ├── policy.go # RoutingPolicy, PolicyMode, ProviderRef🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec.md around lines 319 - 330, The fenced code block that lists the package structure (the block beginning with "internal/llm/router/v2/") is missing a language identifier; update the opening fence from ``` to ```text so the directory tree is treated as plain text (i.e., change the code fence that starts the block containing "internal/llm/router/v2/" to use ```text).
169-177: 💤 Low valueDependency graph 블록에 language identifier 추가 권장
의존성 그래프 diagram이 fenced code block으로 작성되었지만 언어 식별자가 누락되었습니다. ASCII art 형태이므로
text식별자를 추가하면 렌더링이 개선됩니다.제안된 수정
-``` +```text SPEC-GOOSE-LLM-001 (Provider 인터페이스, ProviderCapabilities) └─ SPEC-GOOSE-ROUTER-001 (Smart Routing v1, ProviderRegistry, completed)🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the rest with a brief reason, keep changes minimal, and validate. In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec.md around lines 169 - 177, The fenced dependency graph containing the ASCII art lines beginning with "SPEC-GOOSE-LLM-001 (Provider 인터페이스, ProviderCapabilities)" should have a language identifier added to the opening fence; change the opening "```" to "```text" for that code block (keeping the closing "```") so the diagram (including SPEC-GOOSE-ROUTER-001, SPEC-GOOSE-RATELIMIT-001, SPEC-GOOSE-ERROR-CLASS-001, SPEC-GOOSE-ADAPTER-001/002) renders correctly..moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/acceptance.md (1)
199-202: 💤 Low valueTrace 출력 예시 블록에 language identifier 추가 권장
여러 AC 시나리오의 trace 출력 예시가 fenced code block으로 작성되었지만 언어 식별자가 누락되었습니다.
text또는log식별자를 추가하면 렌더링이 개선되고 markdownlint 경고를 해소할 수 있습니다.해당 위치:
- Line 199-202 (AC-RV2-006)
- Line 231-234 (AC-RV2-007)
- Line 266-269 (AC-RV2-008)
- Line 324-328 (AC-RV2-010)
제안된 수정 (예시)
- ``` + ```text v2:fallback_chain_step_1_rate_limit (anthropic, claude-sonnet-4.6, 120ms) v2:fallback_chain_step_2_success (openai, gpt-4o, 230ms)</details> Also applies to: 231-234, 266-269, 324-328 <details> <summary>🤖 Prompt for AI Agents</summary>Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/acceptance.md around lines 199 -
202, Add a language identifier (e.g., "text" or "log") to the fenced code blocks
that show trace output so markdown renders correctly and linter warnings are
resolved; locate the example blocks referenced by AC IDs AC-RV2-006, AC-RV2-007,
AC-RV2-008, and AC-RV2-010 (the blocks containing lines like
"v2:fallback_chain_step_1_rate_limit (anthropic, claude-sonnet-4.6, 120ms) ...")
and change the opening fences fromtotext (or ```log).</details> </blockquote></details> <details> <summary>.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/plan.md (1)</summary><blockquote> `283-294`: _💤 Low value_ **Fenced code block에 language identifier 추가 권장** plan_complete signal 블록이 언어 식별자 없이 정의되어 있습니다. `yaml` 또는 `text`를 명시하면 렌더링 및 린터 경고를 해소할 수 있습니다. <details> <summary>제안된 수정</summary> ```diff -``` +```yaml <moai>plan_complete</moai> SPEC: SPEC-GOOSE-LLM-ROUTING-V2-001 ``` </details> <details> <summary>🤖 Prompt for AI Agents</summary>Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/plan.md around lines 283 - 294,
The fenced code block containing the plan_complete signal is
missing a language identifier; update the opening fence for that block fromtoyaml (or ```text) so the block is explicitly typed (e.g., change the
triple-backtick that begins the block with a language identifier), ensuring
linters and renderers treat it correctly and preserve the existing block
contents.</details> </blockquote></details> <details> <summary>.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec-compact.md (1)</summary><blockquote> `63-80`: _💤 Low value_ **Table 블록에 language identifier 추가 권장** Capability matrix와 pricing table이 fenced code block으로 작성되었지만 언어 식별자가 누락되었습니다. 테이블 형태이므로 `text` 식별자를 추가하면 렌더링이 개선되고 markdownlint 경고를 해소할 수 있습니다. 해당 위치: - Line 63-80 (Capability matrix) - Line 86-105 (Pricing table) <details> <summary>제안된 수정 (예시)</summary> ```diff -``` +```text prompt_cache func_call vision realtime anthropic ✅ ✅ ✅ ❌ ``` </details> Also applies to: 86-105 <details> <summary>🤖 Prompt for AI Agents</summary>Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec-compact.md around lines 63 -
80, The fenced code blocks containing the capability matrix and the pricing
table currently lack a language identifier; update the opening backticks for
those table blocks (the triple-backtick blocks that wrap the "prompt_cache
func_call vision realtime" matrix and the subsequent pricing table) to include
the text identifier (i.e. changetotext) so markdown rendering and
markdownlint warnings are fixed.</details> </blockquote></details> </blockquote></details> <details> <summary>🤖 Prompt for all review comments with AI agents</summary>Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.Inline comments:
In @.moai/reports/plan-audit/SPEC-GOOSE-MSG-TELEGRAM-001-review-2.md:
- Line 146: 문서의 감사 리포트 작성일이 미래 날짜로 잘못 기록되어 있습니다: 찾아서 "Report written:
2026-05-06" 항목을 현재 기준 실제 작성일인 "Report written: 2026-05-05"로 수정하세요 (파일 내 해당 텍스트을
검색해 교체; 참조 텍스트: "Report written: 2026-05-06").In @.moai/reports/plan-audit/SPEC-GOOSE-SCHEDULER-001-review-1.md:
- Line 25: Update the report's metadata field "status" (currently set as
status: draft) to the current SPEC stateaudit-readyso the MP-3 evidence
line matches the actual SPEC status; locate thestatusentry in the document
header/metadata (thestatus: draftline referenced in the review) and replace
the value withaudit-ready.In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec-compact.md:
- Line 5: The SPEC metadata in spec-compact.md for SPEC-GOOSE-LLM-ROUTING-V2-001
has status=draft which conflicts with spec.md (status=audit-ready) and PR
objectives; update the Meta line for id=SPEC-GOOSE-LLM-ROUTING-V2-001 in
spec-compact.md to status=audit-ready so both files and PR objectives match,
leaving the rest of the Meta fields unchanged and ensure no other compact SPECs
have mismatched statuses.In @.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/spec.md:
- Line 247: The table's code span containing setup|start|status|approve|revoke
is breaking Markdown table parsing because the pipe characters are unescaped;
update the cell that currently showscmd/goose/cmd/telegram.goand the inline
code span to either escape each pipe as\|(i.e.,
setup\|start\|status\|approve\|revoke) or replace the inline backtick code
span with an HTML code element like
setup|start|status|approve|revokeso the table columns render
correctly; search for the exact text "setup|start|status|approve|revoke" in the
spec to locate and fix the issue.- Line 278: The table row for "R2" contains the Telegram reserved-char code span
that currently includes an unescaped pipe character; escape the pipe by changing
'|' to '|' inside that code span (the reserved-char list that includes_ * [ ] ( ) ~ \> # + - = | { } . !`) so the Markdown table column parsing is preserved
and MD056 is avoided—update the code span in the spec.md table row for R2
accordingly.In @.moai/specs/SPEC-GOOSE-SCHEDULER-001/acceptance.md:
- Around line 268-270: Update AC-SCHED-018(a) to remove the ambiguous "or panic"
wording and state a single deterministic outcome: when theFastForwardsymbol
(andErrFastForwardNotAvailable) are gated by//go:build test_only, a
production build must fail at compile/link time (no runtime panic possible).
Ensure the text explicitly says the symbol is not linked in production builds
and that absence causes a compile/link error, and keep the existing check
sentence about verifying file-level gating with//go:build test_only.In @.moai/specs/SPEC-GOOSE-SCHEDULER-001/plan.md:
- Around line 34-40: 파일의 두 개의 펜스드 코드블록(하나는 ASCII 아트 다이어그램 "P1 (cron +
persistence) ... P4 (catchup ...)"을 포함하는 블록, 다른 하나는 "key = fmt.Sprintf..." 예시를
포함하는 블록)에 언어 식별자를 추가하여 markdownlint MD040 경고를 제거하세요; 구체적으로 각각의 시작을text 로
변경해 펜스 코드블록에 language="text"를 지정하면 됩니다.In @.moai/specs/SPEC-GOOSE-SCHEDULER-001/spec-compact.md:
- Line 81: 문구 "다음 단계: plan-auditor 라운드 1 → PASS 시 status
draft → audit-ready"가
현재 문서의 메타데이터(audit-ready)와 모순됩니다; 해당 "다음 단계" 문구를 현재 상태 기준으로 갱신하세요 — 즉 파일 내 "다음
단계" 라인(현재draft → audit-ready표기)을 찾아draft → audit-ready를 현재 메타데이터에 맞는
전환(예:audit-ready → <다음_상태>또는 리포지토리에 정의된 실제 다음 상태명)으로 교체하고, 필요하면 "현재 상태:
audit-ready" 같은 짧은 문구를 추가해 혼동을 없애세요.- Around line 45-47: The compact summary table's AC mappings for the
"Event-Driven (5)" and "State-Driven (3)" rows are inconsistent with the
canonical Coverage Map: update the "Event-Driven (5)" row to list AC-002,
AC-003, AC-004, AC-009, AC-011 and update the "State-Driven (3)" row to list
AC-006, AC-012 so the compact table matches the canonical mapping (ensure AC-003
and AC-006 are moved to the correct rows).
Nitpick comments:
In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/acceptance.md:
- Around line 199-202: Add a language identifier (e.g., "text" or "log") to the
fenced code blocks that show trace output so markdown renders correctly and
linter warnings are resolved; locate the example blocks referenced by AC IDs
AC-RV2-006, AC-RV2-007, AC-RV2-008, and AC-RV2-010 (the blocks containing lines
like "v2:fallback_chain_step_1_rate_limit (anthropic, claude-sonnet-4.6, 120ms)
...") and change the opening fences fromtotext (or ```log).In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/plan.md:
- Around line 283-294: The fenced code block containing the
plan_complete signal is missing a language identifier; update the
opening fence for that block fromtoyaml (or ```text) so the block is
explicitly typed (e.g., change the triple-backtick that begins the block with a
language identifier), ensuring linters and renderers treat it correctly and
preserve the existing block contents.In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec-compact.md:
- Around line 63-80: The fenced code blocks containing the capability matrix and
the pricing table currently lack a language identifier; update the opening
backticks for those table blocks (the triple-backtick blocks that wrap the
"prompt_cache func_call vision realtime" matrix and the subsequent pricing
table) to include the text identifier (i.e. changetotext) so markdown
rendering and markdownlint warnings are fixed.In @.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec.md:
- Around line 347-355: The fenced decision-tree code block that starts with "1.
v1.Route(req) → primary/cheap 결정" is missing a language identifier; update the
opening triple-backtick to include the text language identifier (```text) so the
block renders correctly (affects the decision tree lines referencing
v1.Route(req) and policy.Mode conditions).- Around line 319-330: The fenced code block that lists the package structure
(the block beginning with "internal/llm/router/v2/") is missing a language
identifier; update the opening fence fromtotext so the directory tree
is treated as plain text (i.e., change the code fence that starts the block
containing "internal/llm/router/v2/" to use ```text).- Around line 169-177: The fenced dependency graph containing the ASCII art
lines beginning with "SPEC-GOOSE-LLM-001 (Provider 인터페이스, ProviderCapabilities)"
should have a language identifier added to the opening fence; change the opening
"" to "text" for that code block (keeping the closing "```") so the
diagram (including SPEC-GOOSE-ROUTER-001, SPEC-GOOSE-RATELIMIT-001,
SPEC-GOOSE-ERROR-CLASS-001, SPEC-GOOSE-ADAPTER-001/002) renders correctly.In @.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/plan.md:
- Around line 86-102: Two fenced code blocks (the ASCII diagram block near
"cmd/goose/cmd/telegram.go (cobra: setup/start/...)" and the pseudocode block
beginning "Start(ctx, deps):") are missing a language tag and trigger
markdownlint MD040; update their opening fences fromtotext so the
blocks are explicitly marked as plain text (apply the same change for the second
block around the Start(ctx, deps) pseudocode).In @.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/spec-compact.md:
- Around line 29-46: The fenced code block containing the directory listing (the
block starting with "internal/messaging/telegram/") lacks a language tag and
triggers markdownlint MD040; update that block by adding a language identifier
such as text (i.e., change the opening fence fromtotext) so the snippet
is consistently rendered/highlighted in the spec file
(SPEC-GOOSE-MSG-TELEGRAM-001/spec-compact.md).In @.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/research.md:
- Around line 244-247: Clarify the apparent contradiction in section "9.6 SSRF
방지": update the text around "본 SPEC 은 이런 IP 를 hard block 하지 않음" and the
following sentence about "default blocklist 에 포함하여 explicit allow 필요" to
state clearly that http_fetch will not be universally hard-blocked but that
these IPs are placed in a default blocklist (so calls are blocked by default)
and require an explicit allow action via the PERMISSION-001 Confirmer (include
how the Confirmer indicates scope), and mention examples like 127.0.0.1,
localhost, 169.254.169.254 and private CIDRs to make the intended behavior
unambiguous.</details> <details> <summary>🪄 Autofix (Beta)</summary> Fix all unresolved CodeRabbit comments on this PR: - [ ] <!-- {"checkboxId": "4b0d0e0a-96d7-4f10-b296-3a18ea78f0b9"} --> Push a commit to this branch (recommended) - [ ] <!-- {"checkboxId": "ff5b1114-7d8c-49e6-8ac1-43f82af23a33"} --> Create a new PR with the fixes </details> --- <details> <summary>ℹ️ Review info</summary> <details> <summary>⚙️ Run configuration</summary> **Configuration used**: Path: .coderabbit.yaml **Review profile**: CHILL **Plan**: Pro **Run ID**: `712a5679-cd37-4964-9a19-e35d0c1b0872` </details> <details> <summary>📥 Commits</summary> Reviewing files that changed from the base of the PR and between ed4570186ee316c820a733128dfbea46444c3538 and e2f4123c2b89dc856a0428e9043e0448f637de1b. </details> <details> <summary>📒 Files selected for processing (24)</summary> * `.moai/reports/plan-audit/SPEC-GOOSE-MSG-TELEGRAM-001-review-1.md` * `.moai/reports/plan-audit/SPEC-GOOSE-MSG-TELEGRAM-001-review-2.md` * `.moai/reports/plan-audit/SPEC-GOOSE-SCHEDULER-001-review-1.md` * `.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/acceptance.md` * `.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/plan.md` * `.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/research.md` * `.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec-compact.md` * `.moai/specs/SPEC-GOOSE-LLM-ROUTING-V2-001/spec.md` * `.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/acceptance.md` * `.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/plan.md` * `.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/research.md` * `.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/spec-compact.md` * `.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/spec.md` * `.moai/specs/SPEC-GOOSE-MSG-TELEGRAM-001/status.txt` * `.moai/specs/SPEC-GOOSE-SCHEDULER-001/acceptance.md` * `.moai/specs/SPEC-GOOSE-SCHEDULER-001/plan.md` * `.moai/specs/SPEC-GOOSE-SCHEDULER-001/spec-compact.md` * `.moai/specs/SPEC-GOOSE-SCHEDULER-001/spec.md` * `.moai/specs/SPEC-GOOSE-SCHEDULER-001/status.txt` * `.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/acceptance.md` * `.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/plan.md` * `.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/research.md` * `.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/spec-compact.md` * `.moai/specs/SPEC-GOOSE-TOOLS-WEB-001/spec.md` </details> </details> <!-- This is an auto-generated comment by CodeRabbit for review status -->
- §목적 첫 줄의 GOOSE-AGENT (백틱 외부) 를 AI.GOOSE 로 치환 - scripts/check-brand.sh 로컬 통과 확인 (0 violations) - PR #118 Brand Notation Check FAILURE 해소용 SPEC: SPEC-GOOSE-TOOLS-WEB-001 🗿 MoAI <email@mo.ai.kr>
CodeRabbit review 지적 중 자명한 drift 3건을 spec.md (정본) 기준으로 정정: - SPEC-GOOSE-LLM-ROUTING-V2-001/spec-compact.md:5 Meta status=draft → audit-ready (spec.md L4 정본과 동기화, Critical drift 해소) - .moai/reports/plan-audit/SPEC-GOOSE-MSG-TELEGRAM-001-review-2.md:146 Report written 2026-05-06 → 2026-05-05 (실제 작성일 보정) - .moai/reports/plan-audit/SPEC-GOOSE-SCHEDULER-001-review-1.md:25 MP-3 인용값 status: draft → audit-ready (spec.md L4 정본과 동기화) Cosmetic markdownlint 경고(MD040/MD056), AC mapping 정정, AC-018 단일경로화는 별도 hygiene PR 또는 후속 sync 에서 처리. SPEC: SPEC-GOOSE-LLM-ROUTING-V2-001, SPEC-GOOSE-MSG-TELEGRAM-001, SPEC-GOOSE-SCHEDULER-001 🗿 MoAI <email@mo.ai.kr>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
GOOSE Sprint 1 (Hermes Agent 추월 8주 로드맵의 첫 sprint) plan phase 완료. 4 SPEC 모두 audit-ready (총 79 EARS REQs + 60 ACs).
4 신규 SPECs
전략 결정 (이번 plan phase에서 결정된 사항)
산출물 (24 files, 6,107 insertions)
Test Plan
Next Steps (After Merge)
각 SPEC /moai run 진입 가능. 권장 순서:
관련 SPECs (예정)
🗿 MoAI email@mo.ai.kr
Summary by CodeRabbit
릴리스 노트