a02feaaff7
三件事，都是同一个根因的症状：子 agent 的类型逻辑复制成多份，
于是标准路径上的能力没跟着传过来。
## 1. 抽出 child 包（类型抽象 + 单一登记表）
新增 internal/agent/core/child/：
child.go Kind / Child（嵌入用）/ State / 残余与通道处置
registry.go 单一 Registry + Entry 接口 + Cleanup 骨架
kind.go ChildKind —— 按两条主轴（上下文 / 工具与输出异步）
kinds.go 四种子 agent 的清单与迁移状态
此前 Agent 上有两张互不知情的表（residents / rtChildren），各带一把锁。
实测新增一种子 agent 要改 18 个生产文件，并已因复制而漂移过一次
（StopRTChildren 漏注销入站登记，见 engineering-lessons.md §2.3）。
现在单数与复数停止**共用** Cleanup，那个形态在结构上不可能复现。
## 2. rt 的输出改为异步（用户的原始诉求）
rtEmit 曾直接 dev.Execute(...)，插件侧延迟（TTS/网络编码/字幕渲染）
直接节流模型的 token 消费 —— 「不攒整段」的收益全被吃掉。
新增 outputpump.go（输出通道的异步壳）：
· Push 立即返回，交付由独立消费 goroutine 完成
· 有界队列（64），满则**丢最旧**（符合 rt 的"最新者胜"）
· 丢弃与设备缺失都留痕，否则"用户听不到"没有任何线索
· Close 排空后退出，不吞最后半句
· 设备每片重新解析（插件崩溃会 Unregister，缓存指针会指向死设备）
## 3. 工具钩子统一（本轮最大发现）
实测：StageBeforeToolcall / StageAfterToolcall 全仓只在
task.go:1022/1079/1194，三处都在 runOneTool / stepToolAfter 里；
而 executeToolCallOutcome 内部**没有任何 runStage**。
⇒ rtchild 与工作式子两条路径**完全跳过插件钩子**：不能审计、
不能拒绝、不能改参、不能脱敏。README 承诺的 before_toolcall
在这两条路径上不成立。
新增 toolbatch.go（唯一批执行实现：批并发 + 双钩子 + 顺序落结果），
两种类型切过去。工作式子的"禁输出工具"与"保持串行"用 batchOpts 注入，
不各写一遍执行逻辑。
## 判据（均经变异验证）
TestRTChild_OutputIsAsync 改回同步 ⇒ 红（实测确认）
TestOutputPump_DropsOldestOnFull 改阻塞/无界 ⇒ 红
TestOutputPump_CloseDrains 丢弃队列 ⇒ 红
TestOutputPump_ResolvesDeviceEachTime 缓存指针 ⇒ 红
TestRTChild_ToolsGoThroughPluginStages 绕开钩子 ⇒ 红
TestRTChild_StageCanDenyTool 同上
TestWorkerToolsGoThroughPluginStages 同上（变异实测确认）
TestWorkerToolsStaySerial 改成并发 ⇒ 红
TestNoChildTypeTablesInCore 源码扫描，禁类型专属表回潮
TestRTChild_HasGraphMemory 等 5 条 rt 接入图记忆
## 顺带修复
· rt 接入图记忆（参照驻留子的轻量内核形态）：此前 graphMem() == nil，
memory_recall / memory_commit 在 rt 里全部不可达，每通电话从零开始。
默认只读（通话高频写入会灌满图谱噪声），4 个失败点合并成一个 rollback
闭包（逐点挂必然漏一处 ⇒ 泄 SQLite 句柄）。
· rt 的 PreferredProvider 走候选链（此前裸调 a.provider，优先源无效且零回退）。
## 保留的语义（逐项确认仍在）
驻留子三层压缩：checkContextFull / handleChildContextFull / CompressResident
三种通道处置：ChannelRelease / Suspend / Takeover
残余 keep/drop + harvestResidentMemory
child.Start() 的 7 个 goroutine
验证：build / vet / gofmt / go test -race 全绿；
且在干净检出（git worktree）上单独 build + test 通过。
⚠️ AI-Assisted Programming Notice: Parts of this project's code, documentation, and commit history were generated or modified with AI assistance. Key changes have been human-reviewed, but please evaluate and verify before use.
HologramAgent
中文: README.md
Self-hosted agent kernel. Runs on your own hardware; plugins attach as separate processes.
make build build-cli
./build/homed -data /tmp/ha
API keys go in the web UI at http://localhost:8080. Requires Linux.
The main difference from typical agent frameworks is context handling: instead of truncating on overflow or asking the model to summarize, it scores history by relevance and sinks the low-scoring events into cold context, then distills those into a graph. Two other things differ: startup order is resolved at build time, and plugins launch as soon as their facilities are ready.
Architecture
homed (kernel)
LLM orchestration · memory · retrieval
│
┌──────────────┼──────────────┐
│ │ │
stdio RPC shared mem event ring
control bulk data notify
│ │ │
└──────────────┼──────────────┘
│
plugins (separate processes)
web UI · QQ · CLI · files · browser
The kernel process handles orchestration, memory and retrieval; HTTP, files and devices belong to
plugins (internal/plugin). A crashed plugin is restarted with backoff 1s / 2s / 3s; the 4th
crash within 5 minutes stops automatic restarting pending human intervention
(cmd/homed/kmod_runtime.go).
Context window
internal/agent/core/context.go
Each event gets two vectors on entry: a dense vector sharing one space with media and cold context, and a TF-IDF word vector as fallback. Pruning scores history against the current input and keeps topK, with the most recent N events exempt from scoring.
N defaults to 10, bounded by budget: min(configured, ContextTokens / avg tokens per event).
That bound came from a measurement. At a 50k window with 5800 tokens per event and a 40000 budget, pinning 10 events needed 58k, more than the budget itself. Every prune met its target without freeing space; the budget was gone after two rounds and the task terminated, with recall falling from 44% to 22%. Small windows now tighten automatically; a 1M window uses the configured value.
Pruned events move to the cold-context layer together with the media blocks they hold. Block IDs
stay the same; only the holding layer changes. Retrieval uses coldcontext_query: Peek to
confirm, then Remove the source copy. There is no fetch-and-delete path, since deletion follows
confirmation.
Three layers
internal/agent/core/context.go · internal/memory/coldcontext/ · internal/memory/graph.go
| Layer | Holds | Reaches the model | Cooling |
|---|---|---|---|
| L1 Context | events + blocks | directly, in the system prompt | low relevance sinks |
| L2 Cold Context | events + blocks (same shape as L1) | inserted back into L1 order on recall | ≤2 accesses and 72h idle, then distilled |
| L3 Stable | triples, nodes are blocks | on recall | never leaves |
A block has exactly one holder at any layer, with no reference counting. Images and audio are blocks in L2/L3 with their own vectors, searched in the same space as text.
Scenes
internal/memory/scene_emerge.go
A scene is built from five weighted features: channel 1.0, conversation partner 1.0, tools 0.8,
topic 0.4, time of day 0.2. A scene is created on the second occurrence of the same fingerprint
(minSceneEvidence = 2); one-off interactions don't create one. Join threshold 0.5, recall 0.35.
Recurrence adds 1 to strength; scenes that stop recurring decay by half-life.
Two entry points run in parallel: a plugin declares the scene at an injection point (e.g.
chan:qq/peer:group_1), or fingerprints cluster on their own. A declared scene derives features
only from its own key and does not absorb the full fingerprint. Absorbing it would make the
declared scene match nearly every later turn of that kind at close to 1.0 similarity, and the
clustering path could never grow narrower scenes.
Seven stages
internal/agent/core/stages.go
| Stage | When | Plugins can |
|---|---|---|
on_input |
message arrives | blocklist, rate limit, short-circuit reply |
pre_action |
context ready, before LLM | inject external data |
post_action |
LLM returns text and tool calls | filter, force redirect |
before_toolcall |
before one tool runs | audit, deny, rewrite args |
after_toolcall |
after one tool runs | redact, sort results |
before_output |
final text ready | format adaptation |
after_output |
after sending | statistics |
The inner loop is post_action → [before_toolcall → run → after_toolcall] → post_action. It exits
when the model stops calling tools, all calls are denied, or the limit is reached.
Everything else
- Channels: web UI, CLI, QQ groups and DMs, A2A, ACP in your IDE
- Tools: files, shell, code, browser (real Chrome), calendar, notes, RSS, music
- Modalities: images, video, voice, screenshots, web pages
28 built-in plugins, 89 model-callable tools. External plugins live one per repo under github.com/HologramAgent.
Downloads
Releases ship three variants:
| Variant | Contents | For |
|---|---|---|
| full | homed + waiter + desktop GUI + systemd unit | Single-machine, everything |
| server | homed + waiter + systemd unit | Servers (no desktop environment) |
| client | waiter + desktop GUI | Connecting to a remote HologramAgent |
- Linux:
.deb(amd64/arm64),.rpm(x86_64),.tar.gz - Windows:
HologramAgent_v1.2.0_{Full,Server,Client}_win64.exe(NSIS installer, includes the AGPL license page). Since v1.2.0homedno longer supports native Windows (it relies on fd inheritance and in-segment offset dereferencing), so the installer bootstraps WSL2 and installs the Linux packages inside the distribution the same way a Linux host would. - Portable:
homeagent-bin-<os>_<arch>.tar.gz(homed/waiter/initconfig) - Verification:
SHA256SUMS
The macOS homed requires a native macOS build (CGO + sqlite3), so release packages ship only waiter/initconfig.
Build
make build build-cli # Build daemon + CLI
make test # go test ./...
make install # Install to system
Dependencies: Go 1.25+, CGo (go-sqlite3), Linux.
homedrequires Linux (it relies on fd inheritance and intra-segment offset dereferencing of the shared memory region; seecmd/homed/platform_windows.go). On Windows onlywaiter.exeis built andhomedruns under WSL2 (see Downloads). macOS can buildwaiter/initconfig;homedmust be built on native macOS.
License
AGPL-3.0-only. See LICENSE.
Bundled third-party components keep their own licenses; full texts ship in
/usr/share/doc/hologramagent/licenses/.
Third-party components shipped with the packages
| Component | License | Location |
|---|---|---|
| Chinese-CLIP ViT-B/16 (ONNX artifacts) | Apache-2.0 | /usr/lib/hologram/models/chinese-clip-vit-b16-onnx/ |
ONNX Runtime (libonnxruntime.so) |
MIT | /usr/lib/hologram/onnxruntime/ |
| jieba dictionary (embedded in the binary) | MIT | internal/memory/jiebadict/ |
| Go dependencies (go-sqlite3, gojieba, bubbletea, …) | MIT / BSD-3 / Apache-2.0 | permissive, AGPL-3.0-compatible |
The package metadata declares this package as AGPL-3.0-only.