diff --git a/run.json b/run.json index cbdd26d9c..92fbf3eea 100644 --- a/run.json +++ b/run.json @@ -517,7 +517,7 @@ "kind": "running" }, "status_updated_at": "2026-05-22T18:57:01.494747Z", - "last_event_at": "2026-05-22T18:57:03.691122Z", + "last_event_at": "2026-05-22T18:57:09.504206Z", "pending_control": null, "checkpoints": [ { @@ -558,9 +558,9 @@ "diff": {} }, { - "seq": 0, + "seq": 27, "checkpoint": { - "timestamp": "2026-05-22T18:57:05.238746Z", + "timestamp": "2026-05-22T18:57:09.500755Z", "current_node": "toolchain", "completed_nodes": [ "start", @@ -568,24 +568,88 @@ ], "node_retries": {}, "context_values": { - "internal.run_id": "01KS8GR1P97NYP93VVXWYCZHYZ", + "internal.node_visit_count": 1, "graph.model_stylesheet": "\n * { model: claude-opus-4-7; }\n ", + "internal.thread_id": "start", + "thread.start.current_node": "toolchain", + "current_node": "toolchain", + "internal.retry_count.start": 0, + "failure_class": "", + "internal.run_id": "01KS8GR1P97NYP93VVXWYCZHYZ", + "internal.fidelity": "compact", + "graph.rankdir": "LR", + "failure_signature": "", + "graph.goal": "# Agent Context Observability Events Implementation Plan\n\n> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.\n\n**Goal:** Add durable API/backend events that report loaded memory files, discovered and activated skills, and per-server MCP tool names for agent runs.\n\n**Architecture:** Keep this API-backend scoped. Emit typed `AgentEvent` variants from the existing `fabro-agent` initialization and skill activation paths, convert them through `fabro-workflow` into durable `fabro-types` run events, and document the event contracts. Do not add run projection fields in this pass; consumers can read the event stream/history.\n\n**Tech Stack:** Rust, Serde, Fabro agent/session events, Fabro workflow event conversion, Fabro MCP connection manager, `cargo nextest`.\n\n---\n\n## Scope\n\nImplement these event changes:\n\n- Add `agent.memory.loaded` with memory file paths, byte counts, loaded byte counts, truncation flags, provider profile, total loaded bytes, and budget bytes.\n- Add `agent.skills.discovered` with source directories, provider profile, and sorted skill summaries.\n- Add persisted `agent.skill.activated` for slash skill expansion and successful `use_skill` tool calls.\n- Enrich `agent.mcp.ready` with names-only tool summaries: qualified tool name and original server tool name.\n\nDo not implement ACP-native equivalents in this pass. Do not include memory file contents in any event payload. Do not include MCP tool descriptions or schemas.\n\n## Existing Patterns To Follow\n\n- Read `docs/internal/events-strategy.md` before changing event variants, names, conversion, or progress JSONL behavior.\n- Read `docs/internal/testing-strategy.md` before adding or reorganizing tests.\n- Follow the current `AgentEvent` flow:\n - `lib/crates/fabro-agent/src/types.rs`\n - `lib/crates/fabro-agent/src/session.rs`\n - `lib/crates/fabro-workflow/src/handler/llm/api.rs`\n - `lib/crates/fabro-workflow/src/event/convert.rs`\n - `lib/crates/fabro-workflow/src/event/names.rs`\n - `lib/crates/fabro-types/src/run_event/agent.rs`\n - `lib/crates/fabro-types/src/run_event/mod.rs`\n- Follow Rust import style from `AGENTS.md`: import types by name, import functions through their parent module, and avoid glob imports in production code.\n\n## File Map\n\n- Modify `lib/crates/fabro-types/src/run_event/agent.rs`: add new prop structs and extend `AgentMcpReadyProps`.\n- Modify `lib/crates/fabro-types/src/run_event/mod.rs`: add `EventBody` variants for the new event names.\n- Modify `lib/crates/fabro-agent/src/types.rs`: add internal `AgentEvent` variants, trace output, and noise filtering decisions.\n- Modify `lib/crates/fabro-agent/src/memory.rs`: return memory content plus metadata instead of bare strings.\n- Modify `lib/crates/fabro-agent/src/session.rs`: emit memory, skills, skill activation, and enriched MCP events.\n- Modify `lib/crates/fabro-agent/src/skills.rs`: emit tool-sourced skill activation from `use_skill`.\n- Modify `lib/crates/fabro-mcp/src/connection_manager.rs`: expose or support deterministic names-only tool summaries per server.\n- Modify `lib/crates/fabro-workflow/src/event/convert.rs`: convert new agent events to durable event bodies.\n- Modify `lib/crates/fabro-workflow/src/event/names.rs`: add event names.\n- Modify `lib/crates/fabro-workflow/src/event/events.rs` only if the agent event name mapping also lives there for these variants.\n- Modify `lib/crates/fabro-workflow/src/event/stored_fields.rs` only if a new event needs non-standard stored fields; otherwise rely on existing `Event::Agent` handling.\n- Modify `docs/internal/events.md`: document new event shapes and the richer MCP payload.\n- Add or update tests in `lib/crates/fabro-agent`, `lib/crates/fabro-mcp`, `lib/crates/fabro-types`, and `lib/crates/fabro-workflow`.\n\n---\n\n### Task 1: Add Typed Durable Event Contracts\n\n**Files:**\n- Modify: `lib/crates/fabro-types/src/run_event/agent.rs`\n- Modify: `lib/crates/fabro-types/src/run_event/mod.rs`\n- Test: existing `fabro-types` run event serde tests, or add focused coverage near the existing run event tests.\n\n- [ ] **Step 1: Add agent memory props**\n\nAdd event prop structs with this shape:\n\n```rust\npub struct AgentMemoryLoadedProps {\n pub provider_profile: String,\n pub files: Vec,\n pub total_loaded_bytes: usize,\n pub budget_bytes: usize,\n pub visit: u32,\n}\n\npub struct AgentMemoryFileProps {\n pub path: String,\n pub byte_count: usize,\n pub loaded_bytes: usize,\n pub truncated: bool,\n}\n```\n\n- [ ] **Step 2: Add skill props**\n\nAdd skill discovery and activation props:\n\n```rust\npub struct AgentSkillsDiscoveredProps {\n pub provider_profile: String,\n pub source_dirs: Vec,\n pub skills: Vec,\n pub visit: u32,\n}\n\npub struct AgentSkillSummary {\n pub name: String,\n pub description: String,\n}\n\npub enum AgentSkillActivationSource {\n Slash,\n Tool,\n}\n\npub struct AgentSkillActivatedProps {\n pub skill_name: String,\n pub source: AgentSkillActivationSource,\n pub visit: u32,\n}\n```\n\nUse serde names `slash` and `tool` for `AgentSkillActivationSource`. If a local enum string pattern already exists, follow that pattern.\n\n- [ ] **Step 3: Extend MCP ready props**\n\nExtend `AgentMcpReadyProps` with a backwards-compatible field:\n\n```rust\n#[serde(default, skip_serializing_if = \"Vec::is_empty\")]\npub tools: Vec,\n```\n\nAdd:\n\n```rust\npub struct AgentMcpToolSummary {\n pub name: String,\n pub original_name: String,\n}\n```\n\n- [ ] **Step 4: Add EventBody variants**\n\nAdd `EventBody` variants using these serialized event names:\n\n- `agent.memory.loaded`\n- `agent.skills.discovered`\n- `agent.skill.activated`\n\nKeep existing `agent.mcp.ready` name unchanged and only enrich its props.\n\n- [ ] **Step 5: Add serde tests**\n\nCover:\n\n- New event names serialize to the expected dot names.\n- `AgentSkillActivationSource` serializes as `slash` and `tool`.\n- Old `agent.mcp.ready` JSON without `tools` still deserializes with `tools == []`.\n\n---\n\n### Task 2: Add Internal Agent Events And Conversion\n\n**Files:**\n- Modify: `lib/crates/fabro-agent/src/types.rs`\n- Modify: `lib/crates/fabro-workflow/src/event/convert.rs`\n- Modify: `lib/crates/fabro-workflow/src/event/names.rs`\n- Modify: `lib/crates/fabro-workflow/src/event/events.rs` if needed by the existing name mapping.\n- Test: `lib/crates/fabro-workflow` event conversion tests.\n\n- [ ] **Step 1: Add internal AgentEvent variants**\n\nAdd variants equivalent to:\n\n```rust\nMemoryLoaded {\n provider_profile: String,\n files: Vec,\n total_loaded_bytes: usize,\n budget_bytes: usize,\n}\n\nSkillsDiscovered {\n provider_profile: String,\n source_dirs: Vec,\n skills: Vec,\n}\n\nSkillActivated {\n skill_name: String,\n source: SkillActivationSource,\n}\n\nMcpServerReady {\n server_name: String,\n tool_count: usize,\n tools: Vec,\n}\n```\n\nPrefer small shared internal structs near `AgentEvent` if that matches the existing file organization.\n\n- [ ] **Step 2: Persist skill activation**\n\nDo not classify `SkillActivated` as streaming noise. The existing `SkillExpanded` event is currently filtered before persistence; replace slash expansion emissions with `SkillActivated { source: Slash }` or keep `SkillExpanded` internal-only if removing it would create unnecessary churn.\n\n- [ ] **Step 3: Add trace behavior**\n\nUpdate `AgentEvent::trace` so the new events emit concise tracing summaries:\n\n- memory loaded: profile, file count, total loaded bytes, budget bytes\n- skills discovered: profile, skill count, source dir count\n- skill activated: name and source\n- MCP ready: server, count, and summary count\n\n- [ ] **Step 4: Convert to durable events**\n\nUpdate `fabro-workflow` event conversion so the new agent events map to the new `fabro-types` props and include `visit`.\n\n- [ ] **Step 5: Add conversion tests**\n\nCover each new event with a focused conversion assertion that checks:\n\n- durable event name\n- `visit`\n- core fields\n- no memory content in the converted payload\n\n---\n\n### Task 3: Emit Memory Loaded Metadata\n\n**Files:**\n- Modify: `lib/crates/fabro-agent/src/memory.rs`\n- Modify: `lib/crates/fabro-agent/src/session.rs`\n- Test: relevant `fabro-agent` memory/session tests.\n\n- [ ] **Step 1: Change memory discovery return type**\n\nChange memory discovery from bare `Vec` to a document type carrying both prompt content and event metadata:\n\n```rust\npub struct MemoryDocument {\n pub path: String,\n pub content: String,\n pub byte_count: usize,\n pub loaded_bytes: usize,\n pub truncated: bool,\n}\n```\n\nKeep existing behavior unchanged:\n\n- provider profile filename candidates stay the same\n- root-to-working-dir walk stays the same\n- content dedupe stays the same\n- empty files are skipped\n- total budget remains 32 KiB\n- truncated content keeps the existing truncation marker\n\n- [ ] **Step 2: Preserve prompt assembly behavior**\n\nAdjust session/profile prompt assembly to pass only memory contents where prompt assembly expects memory text. The system prompt should be byte-for-byte equivalent except where existing tests allow non-semantic ordering differences.\n\n- [ ] **Step 3: Emit agent.memory.loaded**\n\nIn `Session::initialize()`, emit `AgentEvent::MemoryLoaded` immediately after memory discovery, before skills and MCP initialization.\n\nEmit the event even when no memory files are loaded. That lets consumers distinguish \"no memory\" from \"not reported.\"\n\n- [ ] **Step 4: Add memory tests**\n\nCover:\n\n- loaded file path appears in event metadata\n- `byte_count` is the original file byte count\n- `loaded_bytes` reflects bytes actually loaded into the prompt budget\n- `truncated` is true only for truncated files\n- event payload never contains memory file contents\n- empty discovery still emits a memory-loaded event with `files == []`\n\n---\n\n### Task 4: Emit Skills Discovered And Skill Activated\n\n**Files:**\n- Modify: `lib/crates/fabro-agent/src/session.rs`\n- Modify: `lib/crates/fabro-agent/src/skills.rs`\n- Test: relevant `fabro-agent` skill/session tests.\n\n- [ ] **Step 1: Emit skills discovered**\n\nAfter `discover_skills(...)`, emit `AgentEvent::SkillsDiscovered` with:\n\n- `provider_profile`\n- `source_dirs`\n- sorted `skills: [{ name, description }]`\n\nEmit the event even when no skills are discovered.\n\n- [ ] **Step 2: Emit slash activation**\n\nWhere slash skill expansion currently emits or creates `SkillExpanded`, emit:\n\n```rust\nAgentEvent::SkillActivated {\n skill_name,\n source: SkillActivationSource::Slash,\n}\n```\n\n- [ ] **Step 3: Emit tool activation**\n\nIn `make_use_skill_tool`, use `ToolContext::emit_agent_event(...)` after a requested skill is found and before returning the skill template. Emit:\n\n```rust\nAgentEvent::SkillActivated {\n skill_name: name.to_string(),\n source: SkillActivationSource::Tool,\n}\n```\n\nDo not emit activation for failed `use_skill` lookups.\n\n- [ ] **Step 4: Add skill tests**\n\nCover:\n\n- discovery event includes all discovered skills sorted by name\n- discovery event includes configured source directories\n- empty discovery emits `skills == []`\n- slash expansion emits `source == slash`\n- successful `use_skill` emits `source == tool`\n- failed `use_skill` does not emit activation\n\n---\n\n### Task 5: Enrich agent.mcp.ready With Names-Only Tool Summaries\n\n**Files:**\n- Modify: `lib/crates/fabro-mcp/src/connection_manager.rs`\n- Modify: `lib/crates/fabro-agent/src/session.rs`\n- Test: relevant `fabro-mcp` or `fabro-agent` MCP tests.\n\n- [ ] **Step 1: Add deterministic tool summaries**\n\nExpose a helper on `McpConnectionManager` or compute in `Session` from `all_tools()`:\n\n- filter tools by `server_name`\n- return qualified tool name as `name`\n- return server-provided tool name as `original_name`\n- sort by qualified `name`\n\n- [ ] **Step 2: Enrich ready emissions**\n\nWhen emitting `AgentEvent::McpServerReady`, include the tool summaries for that server. Keep existing `server_name` and `tool_count`.\n\n- [ ] **Step 3: Add MCP tests**\n\nCover:\n\n- ready event includes only tools from the ready server\n- summaries are sorted by qualified name\n- `name` is the Fabro-qualified MCP tool name\n- `original_name` is the server-provided tool name\n- descriptions and input schemas are not included\n\n---\n\n### Task 6: Update Event Documentation\n\n**Files:**\n- Modify: `docs/internal/events.md`\n\n- [ ] **Step 1: Document new events**\n\nAdd sections for:\n\n- `agent.memory.loaded`\n- `agent.skills.discovered`\n- `agent.skill.activated`\n\nFor `agent.memory.loaded`, explicitly state that file contents are excluded.\n\n- [ ] **Step 2: Update MCP ready docs**\n\nUpdate `agent.mcp.ready` to show:\n\n```json\n{\n \"server_name\": \"github\",\n \"tool_count\": 2,\n \"tools\": [\n {\n \"name\": \"mcp__github__create_issue\",\n \"original_name\": \"create_issue\"\n }\n ],\n \"visit\": 1\n}\n```\n\n- [ ] **Step 3: Record skill event replacement**\n\nIf `agent.skill.expanded` remains in internal code or docs, mark it internal-only or replaced by `agent.skill.activated`.\n\n---\n\n### Task 7: Verify\n\n**Files:**\n- No new files unless test placement requires it.\n\n- [ ] **Step 1: Run focused tests**\n\nRun:\n\n```bash\ncargo nextest run -p fabro-agent -p fabro-workflow -p fabro-types -p fabro-mcp\n```\n\n- [ ] **Step 2: Run formatting**\n\nRun:\n\n```bash\ncargo +nightly-2026-04-14 fmt --all\n```\n\n- [ ] **Step 3: Run clippy for touched crates or workspace**\n\nPrefer the workspace command if time permits:\n\n```bash\ncargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings\n```\n\n- [ ] **Step 4: Final sanity checks**\n\nConfirm:\n\n- memory events never contain file contents\n- skills discovered and memory loaded are emitted even for empty lists\n- skill activation is persisted rather than filtered as streaming noise\n- `agent.mcp.ready` remains backwards-compatible for old events without `tools`\n- docs match the serialized event names and payload shapes\n\n", "command.output": "blob://sha256/fc14b2ba2d770e5cd3169df7a29525c962adfc4cfa3097b9098c63ebd61a748c", + "internal.retry_count.toolchain": 0, + "outcome": "succeeded", + "internal.work_dir": "/home/daytona/workspace/fabro" + }, + "node_outcomes": { + "start": { + "status": "succeeded", + "usage": null + }, + "toolchain": { + "status": "succeeded", + "context_updates": { + "command.output": "blob://sha256/fc14b2ba2d770e5cd3169df7a29525c962adfc4cfa3097b9098c63ebd61a748c" + }, + "notes": "Script completed: command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", + "usage": null + } + }, + "next_node_id": "preflight_compile", + "git_commit_sha": "204456705c23d19d86c10a626219fb27764cfcc9", + "node_visits": { + "toolchain": 1, + "start": 1 + } + }, + "diff": { + "summary": { + "files_changed": 0, + "additions": 0, + "deletions": 0 + } + } + }, + { + "seq": 0, + "checkpoint": { + "timestamp": "2026-05-22T18:59:26.889555Z", + "current_node": "preflight_compile", + "completed_nodes": [ + "start", + "toolchain", + "preflight_compile" + ], + "node_retries": {}, + "context_values": { + "internal.run_id": "01KS8GR1P97NYP93VVXWYCZHYZ", + "thread.toolchain.current_node": "preflight_compile", + "graph.model_stylesheet": "\n * { model: claude-opus-4-7; }\n ", + "command.output": "blob://sha256/12ae32cb1ec02d01eda3581b127c1fee3b0dc53572ed6baf239721a03d82e126", "failure_class": "", "internal.work_dir": "/home/daytona/workspace/fabro", "failure_signature": "", - "internal.thread_id": "start", + "internal.thread_id": "toolchain", + "internal.retry_count.preflight_compile": 0, "internal.node_visit_count": 1, "internal.retry_count.start": 0, "internal.retry_count.toolchain": 0, "internal.fidelity": "compact", "outcome": "succeeded", - "current_node": "toolchain", + "current_node": "preflight_compile", "thread.start.current_node": "toolchain", "graph.goal": "# Agent Context Observability Events Implementation Plan\n\n> **For agentic workers:** REQUIRED SUB-SKILL: Use superpowers:subagent-driven-development (recommended) or superpowers:executing-plans to implement this plan task-by-task. Steps use checkbox (`- [ ]`) syntax for tracking.\n\n**Goal:** Add durable API/backend events that report loaded memory files, discovered and activated skills, and per-server MCP tool names for agent runs.\n\n**Architecture:** Keep this API-backend scoped. Emit typed `AgentEvent` variants from the existing `fabro-agent` initialization and skill activation paths, convert them through `fabro-workflow` into durable `fabro-types` run events, and document the event contracts. Do not add run projection fields in this pass; consumers can read the event stream/history.\n\n**Tech Stack:** Rust, Serde, Fabro agent/session events, Fabro workflow event conversion, Fabro MCP connection manager, `cargo nextest`.\n\n---\n\n## Scope\n\nImplement these event changes:\n\n- Add `agent.memory.loaded` with memory file paths, byte counts, loaded byte counts, truncation flags, provider profile, total loaded bytes, and budget bytes.\n- Add `agent.skills.discovered` with source directories, provider profile, and sorted skill summaries.\n- Add persisted `agent.skill.activated` for slash skill expansion and successful `use_skill` tool calls.\n- Enrich `agent.mcp.ready` with names-only tool summaries: qualified tool name and original server tool name.\n\nDo not implement ACP-native equivalents in this pass. Do not include memory file contents in any event payload. Do not include MCP tool descriptions or schemas.\n\n## Existing Patterns To Follow\n\n- Read `docs/internal/events-strategy.md` before changing event variants, names, conversion, or progress JSONL behavior.\n- Read `docs/internal/testing-strategy.md` before adding or reorganizing tests.\n- Follow the current `AgentEvent` flow:\n - `lib/crates/fabro-agent/src/types.rs`\n - `lib/crates/fabro-agent/src/session.rs`\n - `lib/crates/fabro-workflow/src/handler/llm/api.rs`\n - `lib/crates/fabro-workflow/src/event/convert.rs`\n - `lib/crates/fabro-workflow/src/event/names.rs`\n - `lib/crates/fabro-types/src/run_event/agent.rs`\n - `lib/crates/fabro-types/src/run_event/mod.rs`\n- Follow Rust import style from `AGENTS.md`: import types by name, import functions through their parent module, and avoid glob imports in production code.\n\n## File Map\n\n- Modify `lib/crates/fabro-types/src/run_event/agent.rs`: add new prop structs and extend `AgentMcpReadyProps`.\n- Modify `lib/crates/fabro-types/src/run_event/mod.rs`: add `EventBody` variants for the new event names.\n- Modify `lib/crates/fabro-agent/src/types.rs`: add internal `AgentEvent` variants, trace output, and noise filtering decisions.\n- Modify `lib/crates/fabro-agent/src/memory.rs`: return memory content plus metadata instead of bare strings.\n- Modify `lib/crates/fabro-agent/src/session.rs`: emit memory, skills, skill activation, and enriched MCP events.\n- Modify `lib/crates/fabro-agent/src/skills.rs`: emit tool-sourced skill activation from `use_skill`.\n- Modify `lib/crates/fabro-mcp/src/connection_manager.rs`: expose or support deterministic names-only tool summaries per server.\n- Modify `lib/crates/fabro-workflow/src/event/convert.rs`: convert new agent events to durable event bodies.\n- Modify `lib/crates/fabro-workflow/src/event/names.rs`: add event names.\n- Modify `lib/crates/fabro-workflow/src/event/events.rs` only if the agent event name mapping also lives there for these variants.\n- Modify `lib/crates/fabro-workflow/src/event/stored_fields.rs` only if a new event needs non-standard stored fields; otherwise rely on existing `Event::Agent` handling.\n- Modify `docs/internal/events.md`: document new event shapes and the richer MCP payload.\n- Add or update tests in `lib/crates/fabro-agent`, `lib/crates/fabro-mcp`, `lib/crates/fabro-types`, and `lib/crates/fabro-workflow`.\n\n---\n\n### Task 1: Add Typed Durable Event Contracts\n\n**Files:**\n- Modify: `lib/crates/fabro-types/src/run_event/agent.rs`\n- Modify: `lib/crates/fabro-types/src/run_event/mod.rs`\n- Test: existing `fabro-types` run event serde tests, or add focused coverage near the existing run event tests.\n\n- [ ] **Step 1: Add agent memory props**\n\nAdd event prop structs with this shape:\n\n```rust\npub struct AgentMemoryLoadedProps {\n pub provider_profile: String,\n pub files: Vec,\n pub total_loaded_bytes: usize,\n pub budget_bytes: usize,\n pub visit: u32,\n}\n\npub struct AgentMemoryFileProps {\n pub path: String,\n pub byte_count: usize,\n pub loaded_bytes: usize,\n pub truncated: bool,\n}\n```\n\n- [ ] **Step 2: Add skill props**\n\nAdd skill discovery and activation props:\n\n```rust\npub struct AgentSkillsDiscoveredProps {\n pub provider_profile: String,\n pub source_dirs: Vec,\n pub skills: Vec,\n pub visit: u32,\n}\n\npub struct AgentSkillSummary {\n pub name: String,\n pub description: String,\n}\n\npub enum AgentSkillActivationSource {\n Slash,\n Tool,\n}\n\npub struct AgentSkillActivatedProps {\n pub skill_name: String,\n pub source: AgentSkillActivationSource,\n pub visit: u32,\n}\n```\n\nUse serde names `slash` and `tool` for `AgentSkillActivationSource`. If a local enum string pattern already exists, follow that pattern.\n\n- [ ] **Step 3: Extend MCP ready props**\n\nExtend `AgentMcpReadyProps` with a backwards-compatible field:\n\n```rust\n#[serde(default, skip_serializing_if = \"Vec::is_empty\")]\npub tools: Vec,\n```\n\nAdd:\n\n```rust\npub struct AgentMcpToolSummary {\n pub name: String,\n pub original_name: String,\n}\n```\n\n- [ ] **Step 4: Add EventBody variants**\n\nAdd `EventBody` variants using these serialized event names:\n\n- `agent.memory.loaded`\n- `agent.skills.discovered`\n- `agent.skill.activated`\n\nKeep existing `agent.mcp.ready` name unchanged and only enrich its props.\n\n- [ ] **Step 5: Add serde tests**\n\nCover:\n\n- New event names serialize to the expected dot names.\n- `AgentSkillActivationSource` serializes as `slash` and `tool`.\n- Old `agent.mcp.ready` JSON without `tools` still deserializes with `tools == []`.\n\n---\n\n### Task 2: Add Internal Agent Events And Conversion\n\n**Files:**\n- Modify: `lib/crates/fabro-agent/src/types.rs`\n- Modify: `lib/crates/fabro-workflow/src/event/convert.rs`\n- Modify: `lib/crates/fabro-workflow/src/event/names.rs`\n- Modify: `lib/crates/fabro-workflow/src/event/events.rs` if needed by the existing name mapping.\n- Test: `lib/crates/fabro-workflow` event conversion tests.\n\n- [ ] **Step 1: Add internal AgentEvent variants**\n\nAdd variants equivalent to:\n\n```rust\nMemoryLoaded {\n provider_profile: String,\n files: Vec,\n total_loaded_bytes: usize,\n budget_bytes: usize,\n}\n\nSkillsDiscovered {\n provider_profile: String,\n source_dirs: Vec,\n skills: Vec,\n}\n\nSkillActivated {\n skill_name: String,\n source: SkillActivationSource,\n}\n\nMcpServerReady {\n server_name: String,\n tool_count: usize,\n tools: Vec,\n}\n```\n\nPrefer small shared internal structs near `AgentEvent` if that matches the existing file organization.\n\n- [ ] **Step 2: Persist skill activation**\n\nDo not classify `SkillActivated` as streaming noise. The existing `SkillExpanded` event is currently filtered before persistence; replace slash expansion emissions with `SkillActivated { source: Slash }` or keep `SkillExpanded` internal-only if removing it would create unnecessary churn.\n\n- [ ] **Step 3: Add trace behavior**\n\nUpdate `AgentEvent::trace` so the new events emit concise tracing summaries:\n\n- memory loaded: profile, file count, total loaded bytes, budget bytes\n- skills discovered: profile, skill count, source dir count\n- skill activated: name and source\n- MCP ready: server, count, and summary count\n\n- [ ] **Step 4: Convert to durable events**\n\nUpdate `fabro-workflow` event conversion so the new agent events map to the new `fabro-types` props and include `visit`.\n\n- [ ] **Step 5: Add conversion tests**\n\nCover each new event with a focused conversion assertion that checks:\n\n- durable event name\n- `visit`\n- core fields\n- no memory content in the converted payload\n\n---\n\n### Task 3: Emit Memory Loaded Metadata\n\n**Files:**\n- Modify: `lib/crates/fabro-agent/src/memory.rs`\n- Modify: `lib/crates/fabro-agent/src/session.rs`\n- Test: relevant `fabro-agent` memory/session tests.\n\n- [ ] **Step 1: Change memory discovery return type**\n\nChange memory discovery from bare `Vec` to a document type carrying both prompt content and event metadata:\n\n```rust\npub struct MemoryDocument {\n pub path: String,\n pub content: String,\n pub byte_count: usize,\n pub loaded_bytes: usize,\n pub truncated: bool,\n}\n```\n\nKeep existing behavior unchanged:\n\n- provider profile filename candidates stay the same\n- root-to-working-dir walk stays the same\n- content dedupe stays the same\n- empty files are skipped\n- total budget remains 32 KiB\n- truncated content keeps the existing truncation marker\n\n- [ ] **Step 2: Preserve prompt assembly behavior**\n\nAdjust session/profile prompt assembly to pass only memory contents where prompt assembly expects memory text. The system prompt should be byte-for-byte equivalent except where existing tests allow non-semantic ordering differences.\n\n- [ ] **Step 3: Emit agent.memory.loaded**\n\nIn `Session::initialize()`, emit `AgentEvent::MemoryLoaded` immediately after memory discovery, before skills and MCP initialization.\n\nEmit the event even when no memory files are loaded. That lets consumers distinguish \"no memory\" from \"not reported.\"\n\n- [ ] **Step 4: Add memory tests**\n\nCover:\n\n- loaded file path appears in event metadata\n- `byte_count` is the original file byte count\n- `loaded_bytes` reflects bytes actually loaded into the prompt budget\n- `truncated` is true only for truncated files\n- event payload never contains memory file contents\n- empty discovery still emits a memory-loaded event with `files == []`\n\n---\n\n### Task 4: Emit Skills Discovered And Skill Activated\n\n**Files:**\n- Modify: `lib/crates/fabro-agent/src/session.rs`\n- Modify: `lib/crates/fabro-agent/src/skills.rs`\n- Test: relevant `fabro-agent` skill/session tests.\n\n- [ ] **Step 1: Emit skills discovered**\n\nAfter `discover_skills(...)`, emit `AgentEvent::SkillsDiscovered` with:\n\n- `provider_profile`\n- `source_dirs`\n- sorted `skills: [{ name, description }]`\n\nEmit the event even when no skills are discovered.\n\n- [ ] **Step 2: Emit slash activation**\n\nWhere slash skill expansion currently emits or creates `SkillExpanded`, emit:\n\n```rust\nAgentEvent::SkillActivated {\n skill_name,\n source: SkillActivationSource::Slash,\n}\n```\n\n- [ ] **Step 3: Emit tool activation**\n\nIn `make_use_skill_tool`, use `ToolContext::emit_agent_event(...)` after a requested skill is found and before returning the skill template. Emit:\n\n```rust\nAgentEvent::SkillActivated {\n skill_name: name.to_string(),\n source: SkillActivationSource::Tool,\n}\n```\n\nDo not emit activation for failed `use_skill` lookups.\n\n- [ ] **Step 4: Add skill tests**\n\nCover:\n\n- discovery event includes all discovered skills sorted by name\n- discovery event includes configured source directories\n- empty discovery emits `skills == []`\n- slash expansion emits `source == slash`\n- successful `use_skill` emits `source == tool`\n- failed `use_skill` does not emit activation\n\n---\n\n### Task 5: Enrich agent.mcp.ready With Names-Only Tool Summaries\n\n**Files:**\n- Modify: `lib/crates/fabro-mcp/src/connection_manager.rs`\n- Modify: `lib/crates/fabro-agent/src/session.rs`\n- Test: relevant `fabro-mcp` or `fabro-agent` MCP tests.\n\n- [ ] **Step 1: Add deterministic tool summaries**\n\nExpose a helper on `McpConnectionManager` or compute in `Session` from `all_tools()`:\n\n- filter tools by `server_name`\n- return qualified tool name as `name`\n- return server-provided tool name as `original_name`\n- sort by qualified `name`\n\n- [ ] **Step 2: Enrich ready emissions**\n\nWhen emitting `AgentEvent::McpServerReady`, include the tool summaries for that server. Keep existing `server_name` and `tool_count`.\n\n- [ ] **Step 3: Add MCP tests**\n\nCover:\n\n- ready event includes only tools from the ready server\n- summaries are sorted by qualified name\n- `name` is the Fabro-qualified MCP tool name\n- `original_name` is the server-provided tool name\n- descriptions and input schemas are not included\n\n---\n\n### Task 6: Update Event Documentation\n\n**Files:**\n- Modify: `docs/internal/events.md`\n\n- [ ] **Step 1: Document new events**\n\nAdd sections for:\n\n- `agent.memory.loaded`\n- `agent.skills.discovered`\n- `agent.skill.activated`\n\nFor `agent.memory.loaded`, explicitly state that file contents are excluded.\n\n- [ ] **Step 2: Update MCP ready docs**\n\nUpdate `agent.mcp.ready` to show:\n\n```json\n{\n \"server_name\": \"github\",\n \"tool_count\": 2,\n \"tools\": [\n {\n \"name\": \"mcp__github__create_issue\",\n \"original_name\": \"create_issue\"\n }\n ],\n \"visit\": 1\n}\n```\n\n- [ ] **Step 3: Record skill event replacement**\n\nIf `agent.skill.expanded` remains in internal code or docs, mark it internal-only or replaced by `agent.skill.activated`.\n\n---\n\n### Task 7: Verify\n\n**Files:**\n- No new files unless test placement requires it.\n\n- [ ] **Step 1: Run focused tests**\n\nRun:\n\n```bash\ncargo nextest run -p fabro-agent -p fabro-workflow -p fabro-types -p fabro-mcp\n```\n\n- [ ] **Step 2: Run formatting**\n\nRun:\n\n```bash\ncargo +nightly-2026-04-14 fmt --all\n```\n\n- [ ] **Step 3: Run clippy for touched crates or workspace**\n\nPrefer the workspace command if time permits:\n\n```bash\ncargo +nightly-2026-04-14 clippy --workspace --all-targets -- -D warnings\n```\n\n- [ ] **Step 4: Final sanity checks**\n\nConfirm:\n\n- memory events never contain file contents\n- skills discovered and memory loaded are emitted even for empty lists\n- skill activation is persisted rather than filtered as streaming noise\n- `agent.mcp.ready` remains backwards-compatible for old events without `tools`\n- docs match the serialized event names and payload shapes\n\n", "graph.rankdir": "LR" }, "node_outcomes": { + "start": { + "status": "succeeded", + "usage": null + }, "toolchain": { "status": "succeeded", "context_updates": { @@ -594,15 +658,20 @@ "notes": "Script completed: command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", "usage": null }, - "start": { + "preflight_compile": { "status": "succeeded", + "context_updates": { + "command.output": "blob://sha256/12ae32cb1ec02d01eda3581b127c1fee3b0dc53572ed6baf239721a03d82e126" + }, + "notes": "Script completed: cargo check -q --workspace 2>&1", "usage": null } }, - "next_node_id": "preflight_compile", + "next_node_id": "preflight_lint", "node_visits": { "start": 1, - "toolchain": 1 + "toolchain": 1, + "preflight_compile": 1 } }, "diff": {} @@ -629,22 +698,22 @@ "superseded_by": null, "pending_interviews": {}, "stages": { - "toolchain@1": { - "first_event_seq": 20, + "preflight_compile@1": { + "first_event_seq": 30, "prompt": null, "response": null, "completion": null, "provider_used": null, "diff": null, "script_invocation": { - "script": "command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", - "command": "exec 2>&1\ncommand -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", + "script": "cargo check -q --workspace 2>&1", + "command": "exec 2>&1\ncargo check -q --workspace 2>&1", "language": "shell" }, "script_timing": null, "parallel_results": null, "output": null, - "started_at": "2026-05-22T18:57:03.691098Z", + "started_at": "2026-05-22T18:57:09.503375Z", "handler": "command", "usage": { "input_tokens": 0, @@ -656,6 +725,54 @@ }, "state": "running" }, + "toolchain@1": { + "first_event_seq": 20, + "prompt": null, + "response": null, + "completion": { + "outcome": "succeeded", + "notes": "Script completed: command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", + "failure_reason": null, + "timestamp": "2026-05-22T18:57:05.237508Z" + }, + "provider_used": null, + "diff": null, + "script_invocation": { + "script": "command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", + "command": "exec 2>&1\ncommand -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", + "language": "shell" + }, + "script_timing": { + "output": "blob://sha256/fc14b2ba2d770e5cd3169df7a29525c962adfc4cfa3097b9098c63ebd61a748c", + "exit_code": 0, + "duration_ms": 1539, + "termination": "exited", + "output_bytes": 36, + "live_streaming": true + }, + "parallel_results": null, + "output": null, + "output_bytes": 36, + "live_streaming": true, + "termination": "exited", + "started_at": "2026-05-22T18:57:03.691098Z", + "handler": "command", + "timing": { + "wall_time_ms": 1546, + "inference_time_ms": 0, + "tool_time_ms": 0, + "active_time_ms": 0 + }, + "usage": { + "input_tokens": 0, + "output_tokens": 0, + "total_tokens": 0, + "reasoning_tokens": 0, + "cache_read_tokens": 0, + "cache_write_tokens": 0 + }, + "state": "succeeded" + }, "start@1": { "first_event_seq": 16, "prompt": null, diff --git a/stages/002-toolchain@1/output.log b/stages/002-toolchain@1/output.log new file mode 100644 index 000000000..4e86d161d --- /dev/null +++ b/stages/002-toolchain@1/output.log @@ -0,0 +1 @@ +blob://sha256/fc14b2ba2d770e5cd3169df7a29525c962adfc4cfa3097b9098c63ebd61a748c \ No newline at end of file diff --git a/stages/002-toolchain@1/script_timing.json b/stages/002-toolchain@1/script_timing.json new file mode 100644 index 000000000..a6241908d --- /dev/null +++ b/stages/002-toolchain@1/script_timing.json @@ -0,0 +1,8 @@ +{ + "output": "blob://sha256/fc14b2ba2d770e5cd3169df7a29525c962adfc4cfa3097b9098c63ebd61a748c", + "exit_code": 0, + "duration_ms": 1539, + "termination": "exited", + "output_bytes": 36, + "live_streaming": true +} \ No newline at end of file diff --git a/stages/002-toolchain@1/status.json b/stages/002-toolchain@1/status.json new file mode 100644 index 000000000..5540db189 --- /dev/null +++ b/stages/002-toolchain@1/status.json @@ -0,0 +1,6 @@ +{ + "outcome": "succeeded", + "notes": "Script completed: command -v cargo >/dev/null || { curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh -s -- -y && sudo ln -sf $HOME/.cargo/bin/* /usr/local/bin/; }; cargo --version 2>&1", + "failure_reason": null, + "timestamp": "2026-05-22T18:57:05.237508Z" +} \ No newline at end of file diff --git a/stages/003-preflight_compile@1/script_invocation.json b/stages/003-preflight_compile@1/script_invocation.json new file mode 100644 index 000000000..d3abb832f --- /dev/null +++ b/stages/003-preflight_compile@1/script_invocation.json @@ -0,0 +1,5 @@ +{ + "script": "cargo check -q --workspace 2>&1", + "command": "exec 2>&1\ncargo check -q --workspace 2>&1", + "language": "shell" +} \ No newline at end of file