Why: the dual-speed entry hint is advisory only; real runs (MiniMax, 3 attachments, no text) end with a clarifying question and zero delegation, so background cards and result backfill never start.
Evidence: new unit tests in runmgr/material_test.go and runtime/material_flow_test.go all pass; full go test ./... exit 0 (local, SELF).
Impact: attachment-carrying runs now guarantee a material-organization background task (pre-spawn on attachment-only/heavy sends, stop-hook backstop otherwise when the model did not delegate); subagent receives inline parsed text with budget plus knowledge_search guidance; content-key dedup prevents duplicate spawns (in-flight, completed, resend); model-authored delegate_task gains a materials appendix; settings runtime.materialIntake (default on) and runtime.materialIntakeChars added.
Refs: Dev task .task/feature-material-intake-flow.md I1; user ruling 2026-10-11.
Why: after material organization completes, weak models report and end the turn without offering the user a choice of formal deliverables, and even after a user picks one the model may close the turn without starting the background task.
Evidence: new tests (notice-choice chain, deliverable backstop, free-input path, invalid args, notice contract) pass; full go test ./... exit 0 (local, SELF).
Impact: new confirm_choice tool issues a HITL choice card and routes the answer into a hard delegate instruction; material notices instruct the model to report then offer choices; stop hook adds two one-shot backstops (choice not offered in material notice runs; choice made but no background delegation).
Refs: Dev task .task/feature-material-intake-flow.md I2.
Why: at the 3-per-conversation cap the guaranteed material task was dropped silently, and long deliverables stayed in chat because cardflow defaulted off (user ruling 2026-10-11).
Evidence: new queue lifecycle test (queued card, hard-reject for model path, FIFO start on release, card promote) passes; full go test ./... exit 0 (local, SELF).
Impact: queueable system spawns create a queued progress card at capacity and start FIFO when a slot frees (memory-only queue; restart limitation noted); model delegate_task still hard-rejects; ui.cardflow default true (EffectNewRun).
Refs: Dev task .task/feature-material-intake-flow.md I3.
Blocking a user prevents them from interacting with repositories, such as opening or commenting on pull requests or issues. Learn more about blocking a user.
feature/material-intake-flow 任务卡
任务主记录：<PR 链接，交付前填写>；状态：进行中；owner：weichen
基线与授权
权威仓：ssh://git@git-ssh.u1680832.nyat.app:31411/weichen/Agent-System-Dev.git；目标分支：main；R0（initial_base_sha）= 555fb2162fe47e13835c592013bc3a69e46239e2
工作区：/home/weichen/projects/Agent-System-Dev-wt-feature-material-intake-flow（worktree）；分支：feature/material-intake-flow
授权与交付终点：用户 2026-10-11 会话批准本卡「方案与计划」节所载方案（含四项范围裁定），
授权实施 I1–I4 增量、对应测试与真实模型/浏览器验收、Backend/UI/Frontend/Dev PR 创建与合入、
任务卡闭合；高后果操作仍走双钥匙。
交付模式：manual（逐阶段人工评审）；实施增量 I1–I4 每增量出口人工评审
自动推进授权：无（阶段间评审未授权连续推进）；授权范围：阶段内实施/自检/缺陷修复；
授权来源：用户会话计划批准 2026-10-11；授权终点：I4 验收通过；停止条件：同仓规
当前阶段：正式测试与验收（待人工评审）；当前增量：I4 完成
流程依据：仓库根目录 policy/delivery.md（归档后仍按仓库根目录定位）
目标与非目标
目标：把快慢双速改为默认工作方式，补齐「上传附件直接发送」的完整闭环：系统确定性启动
材料整理后台任务（附件即保证）→ 材料完整传递给子任务 → 卡片即时展示/过程回填 →
整理完成后待确认交付物选择卡 → 选定后后台任务必启动（模型委派 + stop 兜底）且不重复派发；
容量满时系统任务排队展示；
ui.cardflow默认改为开。验收主用例：三个附件、无文字、直接发送（用户 2026-10-11 实测暴露 22 秒 0 卡 0 委派）。
非目标：OCR/扫描件解析（MinerU 层）；无附件通用意图的自动拆分兜底（用户未选）；
agent-native 内核契约改动；设计系统既有约束不动。
关联方向与规划
用户 2026-10-11 会话中逐项裁定默认行为规则并批准方案，方向以本任务卡为凭；
若后续演化为长期规划再按 policy/delivery.md「长期方向」节建主记录）
需求分析
方案与计划
cardflow 默认开）。范围裁定映射：预启动条件=无文字或材料总量≥runtime.materialIntakeChars
（默认 12000 字符）；材料键=sha256(排序后 filename+字节数+解析字符数)；
材料子代理=代码内 fallback spec agent.sub.material（Model="chat"，
tools=fs_read_file/fs_list_dir/knowledge_search）；去重=内存活跃键集+卡片 KV 材料键扫描
（重启安全）；排队=仅系统派发任务，会话级 FIFO，decBackground 释放按序启动。
startClaimedRun 接线、turn-0 提醒改写、stop 兜底、键级防重复派发；对应 R1/R2/R3/R4/R6 部分）
→ I2 待确认交付物链路（通知回合契约、选择回灌、选择后 stop 兜底、默认选择卡应急；
对应 R5/R6）→ I3 容量排队+cardflow 默认开+UI/Frontend 跨仓+Dev pin 级联（对应 R7/R8/R3）
→ I4 正式验收（真实模型前后对照 ≥3 轮 MiniMax + ≥2 轮主推组合；浏览器端到端；对应 R9 及全部）。
I4 按政策 2026-10-09 裁定做真实模型任务级前后对照（BEFORE=当前 main 补跑 1 轮+用户已录现象；
AFTER=新路径多轮），浏览器端到端走 8788 评审环境；make test 组合（含 pin 级联）。
阶段评审
16ed7b2/ ui bfecf8b / Dev 156c119+探索试验与迭代
同步记录