recallnest 1.4.1

package/eval/continuity/memory-seeds.json

@@ -0,0 +1,26 @@
+[
+  {
+    "text": "RecallNest is the shared memory layer for Claude Code, Codex, and Gemini CLI.",
+    "category": "entities",
+    "scope": "recallnest",
+    "importance": 0.92,
+    "tags": ["recallnest", "continuity", "project"],
+    "canonicalKey": "entities:recallnest:shared-memory-layer"
+  },
+  {
+    "text": "RecallNest continuity revolves around three primitives: store_memory for durable facts, checkpoint_session for live handoff state, and resume_context for startup recovery.",
+    "category": "entities",
+    "scope": "recallnest",
+    "importance": 0.89,
+    "tags": ["recallnest", "continuity", "architecture"],
+    "canonicalKey": "entities:recallnest:continuity-primitives"
+  },
+  {
+    "text": "RecallNest exposes continuity through both HTTP API and MCP, so different agents can share one memory layer and the same handoff model.",
+    "category": "entities",
+    "scope": "recallnest",
+    "importance": 0.84,
+    "tags": ["recallnest", "api", "mcp"],
+    "canonicalKey": "entities:recallnest:api-mcp-surface"
+  }
+]

package/eval/continuity/pattern-seeds.json

@@ -0,0 +1,100 @@
+[
+  {
+    "title": "Cross-window continuity handoff",
+    "trigger": "When opening a fresh terminal window for the same project or after context loss",
+    "steps": [
+      "Call resume_context before coding.",
+      "Read stable context and the latest checkpoint before planning work.",
+      "Continue from the recovered state instead of restating the whole project."
+    ],
+    "outcome": "A fresh window starts with the right project background and fewer manual reminders.",
+    "scope": "recallnest",
+    "tools": ["resume_context", "latest_checkpoint"],
+    "tags": ["continuity", "handoff"],
+    "importance": 0.92
+  },
+  {
+    "title": "Checkpoint before switching windows",
+    "trigger": "When leaving an active coding window that will likely be resumed later",
+    "steps": [
+      "Summarize the current task, decisions, open loops, and next actions.",
+      "Write checkpoint_session with the current sessionId or shared scope.",
+      "Only then close the window or switch terminals."
+    ],
+    "outcome": "The next window can recover active work state without depending on raw transcript recall.",
+    "scope": "recallnest",
+    "tools": ["checkpoint_session"],
+    "tags": ["continuity", "checkpoint"],
+    "importance": 0.9
+  },
+  {
+    "title": "Promote recurring continuity workflow",
+    "trigger": "When the same startup or handoff workflow repeats across multiple windows",
+    "steps": [
+      "Rewrite the workflow as title, trigger, ordered steps, and expected outcome.",
+      "Store it with store_workflow_pattern or POST /v1/pattern.",
+      "Prefer the stored durable pattern over temporary fallback rules in future windows."
+    ],
+    "outcome": "Continuity logic gradually moves from built-in fallback to reusable durable patterns.",
+    "scope": "recallnest",
+    "tools": ["store_workflow_pattern", "/v1/pattern"],
+    "tags": ["continuity", "promotion"],
+    "importance": 0.88
+  },
+  {
+    "title": "Recall before repo exploration",
+    "trigger": "When a fresh window continues an existing project and startup context still looks sparse",
+    "steps": [
+      "Call resume_context before reading local files or docs.",
+      "If stable context is still thin, run search_memory with the project name and task nouns.",
+      "Only after recall is established, inspect the repo and continue implementation."
+    ],
+    "outcome": "Fresh windows recover both stable background and task-specific memory before local exploration drifts.",
+    "scope": "recallnest",
+    "tools": ["resume_context", "search_memory"],
+    "tags": ["continuity", "startup", "retrieval"],
+    "importance": 0.91
+  },
+  {
+    "title": "Scoped project continuity recall",
+    "trigger": "When continuing a named project that already has a shared scope or project key",
+    "steps": [
+      "Call resume_context with the project scope when it is known.",
+      "Prefer scoped durable entities, patterns, and cases before falling back to global memory.",
+      "Use search_memory only after scoped continuity results are reviewed."
+    ],
+    "outcome": "Project-specific context appears before unrelated global memories in fresh windows.",
+    "scope": "recallnest",
+    "tools": ["resume_context", "latest_checkpoint", "search_memory"],
+    "tags": ["continuity", "scope", "project"],
+    "importance": 0.89
+  },
+  {
+    "title": "RecallNest MCP transport rollout",
+    "trigger": "When RecallNest continuity work touches MCP transport wiring under project scope",
+    "steps": [
+      "Call resume_context with project:recallnest before transport changes.",
+      "Review scoped patterns and cases for prior transport fixes before editing adapters.",
+      "Keep transport rollout notes scoped to RecallNest so other projects cannot bleed into startup context."
+    ],
+    "outcome": "RecallNest transport changes recover project-specific continuity without leaking neighboring project rollout notes.",
+    "scope": "project:recallnest",
+    "tools": ["resume_context", "search_memory", "eval:continuity"],
+    "tags": ["continuity", "scope", "transport", "mcp"],
+    "importance": 0.87
+  },
+  {
+    "title": "Telegram bridge MCP transport rollout",
+    "trigger": "When Telegram bridge relay work touches MCP transport wiring under project scope",
+    "steps": [
+      "Call resume_context with project:telegram-bridge before bridge transport edits.",
+      "Review bridge-specific rollout notes before changing relay adapters.",
+      "Keep bridge transport rollout guidance isolated from other projects."
+    ],
+    "outcome": "Telegram bridge transport changes recover bridge-specific continuity without contaminating other project scopes.",
+    "scope": "project:telegram-bridge",
+    "tools": ["resume_context", "search_memory"],
+    "tags": ["continuity", "scope", "transport", "bridge"],
+    "importance": 0.74
+  }
+]

package/eval/continuity/results-2026-03-16.md

@@ -0,0 +1,117 @@
+# RecallNest Continuity Eval
+
+- Generated: 2026-03-16T03:35:14.079Z
+- Cases: 5
+- Passed: 5/5
+- Average score: 94.0%
+
+| Case | Profile | Score | Pass | Stable | Patterns | Cases | Checkpoint |
+|------|---------|-------|------|--------|----------|-------|------------|
+| writing_style_continuity | writing | 100% | yes | 3 | 1 | 0 | no |
+| visual_style_continuity | default | 100% | yes | 3 | 0 | 0 | no |
+| project_context_continuity | default | 100% | yes | 1 | 2 | 1 | no |
+| workflow_pattern_gap_check | default | 70% | yes | 3 | 3 | 1 | no |
+| recallnest_checkpoint_continuity | default | 100% | yes | 3 | 1 | 1 | yes |
+
+## Case Notes
+
+### writing_style_continuity
+- Task: 继续写我的AI小木屋的文章
+- Score: 100%
+- Pass: yes
+- Stable items: 3
+- Pattern items: 1
+- Case items: 0
+- Checkpoint present: no
+- Stable preview: Task focus: 小木屋的文章 | Pinned: 用户稳定偏好口语化、不端着、可以自嘲但不说教。表达应像朋友聊天，避免居高临下、用力过猛或碎嘴。 | Pinned: 用户常用视觉风格是手绘涂鸦风加高对比撞色；在内容包装和配图生成时，应优先沿用这一审美方向，除非任务目标明确要求其他风格。
+- Pattern preview: Workflow pattern: Promote recurring continuity workflow Use when: When the same startup or handoff workflow repeats a...
+- Case preview: -
+- Checkpoint summary: -
+- Matched stable any: 口语化, 不端着
+- Matched stable all: -
+- Matched patterns any: -
+- Matched cases any: -
+- Matched checkpoint any: -
+- Forbidden matches: -
+- Notes: Fresh windows should carry over stable writing-style preferences.
+
+### visual_style_continuity
+- Task: 给文章做封面和配图
+- Score: 100%
+- Pass: yes
+- Stable items: 3
+- Pattern items: 0
+- Case items: 0
+- Checkpoint present: no
+- Stable preview: Task focus: 给文章做封面和配图 | Pinned: 用户常用视觉风格是手绘涂鸦风加高对比撞色；在内容包装和配图生成时，应优先沿用这一审美方向，除非任务目标明确要求其他风格。 | Pinned: 用户稳定偏好口语化、不端着、可以自嘲但不说教。表达应像朋友聊天，避免居高临下、用力过猛或碎嘴。
+- Pattern preview: -
+- Case preview: -
+- Checkpoint summary: -
+- Matched stable any: 手绘涂鸦, 高对比撞色
+- Matched stable all: -
+- Matched patterns any: -
+- Matched cases any: -
+- Matched checkpoint any: -
+- Forbidden matches: -
+- Notes: Visual direction should survive without the user restating it.
+
+### project_context_continuity
+- Task: 一起继续排查 RecallNest 的连续性问题
+- Score: 100%
+- Pass: yes
+- Stable items: 1
+- Pattern items: 2
+- Case items: 1
+- Checkpoint present: no
+- Stable preview: Task focus: recallnest
+- Pattern preview: Workflow pattern: Cross-window continuity handoff Use when: When opening a fresh terminal window for the same project... | Workflow pattern: Promote recurring continuity workflow Use when: When the same startup or handoff workflow repeats a...
+- Case preview: 我先看真实召回密度，确认问题是不是出在 `scope: recallnest` 太窄。如果当前库里几乎没有这个 scope 的 durable memory，就该让 `resume_context` 更积极地退回到项目名驱动的 glo...
+- Checkpoint summary: -
+- Matched stable any: RecallNest
+- Matched stable all: -
+- Matched patterns any: -
+- Matched cases any: -
+- Matched checkpoint any: -
+- Forbidden matches: -
+- Notes: Fresh windows should at least recover the active project context.
+
+### workflow_pattern_gap_check
+- Task: 开个新窗口继续做同一个终端项目
+- Score: 70%
+- Pass: yes
+- Stable items: 3
+- Pattern items: 3
+- Case items: 1
+- Checkpoint present: no
+- Stable preview: Task focus: 开个新窗口继续做同一个终端项目 | Pinned: 用户稳定偏好口语化、不端着、可以自嘲但不说教。表达应像朋友聊天，避免居高临下、用力过猛或碎嘴。 | Pinned: 用户偏好先本地测通再上 Bot，避免烧配额；代码改动完成后要求 commit + push，避免本地与 GitHub 版本漂移。
+- Pattern preview: Workflow pattern: Cross-window continuity handoff Use when: When opening a fresh terminal window for the same project... | Workflow pattern: Promote recurring continuity workflow Use when: When the same startup or handoff workflow repeats a... | Workflow pattern: Checkpoint before switching windows Use when: When leaving an active coding window that will likely...
+- Case preview: Case: Three-terminal continuity trigger validation Problem: Claude Code, Codex, and Gemini CLI were configured with M...
+- Checkpoint summary: -
+- Matched stable any: -
+- Matched stable all: -
+- Matched patterns any: resume_context, checkpoint
+- Matched cases any: -
+- Matched checkpoint any: -
+- Forbidden matches: -
+- Notes: This is a stretch case. It should start passing as more workflow patterns are promoted into durable memory.
+
+### recallnest_checkpoint_continuity
+- Task: 继续 recallnest
+- Score: 100%
+- Pass: yes
+- Stable items: 3
+- Pattern items: 1
+- Case items: 1
+- Checkpoint present: yes
+- Stable preview: Checkpoint focus: RecallNest continuity layer — 状态交接给下个窗口 | Checkpoint summary: RecallNest Phase 3/4 并行推进中。核心三件套 store_memory/checkpoint_session/resume_context 已上线三端 MCP，但 resum... | Checkpoint decision: Phase 3/4 并行推进，continuity layer 是当前核心方向
+- Pattern preview: Workflow pattern: Promote recurring continuity workflow Use when: When the same startup or handoff workflow repeats a...
+- Case preview: Case: RecallNest sparse startup context cleanup Problem: resume_context was returning noisy transcript fragments and ...
+- Checkpoint summary: RecallNest Phase 3/4 并行推进中。核心三件套 store_memory/checkpoint_session/resume_context 已上线三端 MCP，但 resume_context compose 质量是最大短板——实测只返回 3 条 stable context、0 patter...
+- Matched stable any: -
+- Matched stable all: -
+- Matched patterns any: -
+- Matched cases any: -
+- Matched checkpoint any: Phase 3, resume_context, store_memory
+- Forbidden matches: -
+- Notes: When a project checkpoint exists, resume_context should prefer checkpoint-led continuity over noisy transcript fragments.
+