feat: Structured compaction via unified ledger

README.md

@@ -135,29 +135,55 @@ Each task gets its own implement→review loop:
 
 ### 4. Session Continuity
 
-Maintain context across long sessions and context clears with the ledger system:
+Maintain context across long sessions and context clears with structured compaction:
 
 #### Ledger System
 
-The **continuity ledger** captures essential session state:
+The **continuity ledger** serves as both session state and compaction summary. Based on [Factory.ai's structured compaction research](https://factory.ai/blog/context-compression), which found that structured summarization with deterministic file tracking retains more useful context.
 
 ```
 /ledger
 ```
 
 Creates/updates `thoughts/ledgers/CONTINUITY_{session-name}.md` with:
-- Goal and constraints
-- Key decisions with rationale
-- Current state (Done/Now/Next)
-- Working set (branch, key files)
 
-**Auto-injection:** When starting a session, the most recent ledger is automatically injected into the system prompt.
+```markdown
+# Session: {name}
+Updated: {timestamp}
+
+## Goal
+## Constraints
+## Progress
+### Done
+- [x] {Completed items}
+### In Progress
+- [ ] {Current work}
+### Blocked
+- {Issues, if any}
+## Key Decisions
+- **{Decision}**: {Rationale}
+## Next Steps
+1. {Ordered list}
+## File Operations
+### Read
+- `{paths read since last compaction}`
+### Modified
+- `{paths written/edited since last compaction}`
+## Critical Context
+- {Data, examples, references needed to continue}
+```
+
+**Key features:**
+
+- **Iterative merging** - Updates preserve existing information, adding new progress rather than regenerating from scratch
+- **Deterministic file tracking** - Read/write/edit operations tracked automatically via tool call interception, not LLM extraction
+- **Auto-injection** - Most recent ledger injected into system prompt on session start
 
 **Auto-clear:** At 80% context usage, the system automatically:
-1. Updates the ledger
-2. Creates a handoff document
+1. Captures file operations tracked since last clear
+2. Updates ledger with current state (iterative merge with previous)
 3. Clears the session
-4. Injects the ledger into the fresh context
+4. Injects the updated ledger into fresh context
 
 #### Artifact Search
 
@@ -170,26 +196,17 @@ Search past work to find relevant precedent:
 
 Searches across:
 - Ledgers (`thoughts/ledgers/`)
-- Handoffs (`thoughts/shared/handoffs/`)
 - Plans (`thoughts/shared/plans/`)
 
 **Auto-indexing:** Artifacts are automatically indexed when created.
 
-#### Handoff
-
-Save/resume session state for continuity:
-
-- `handoff-creator`: Save current session (reads ledger for context)
-- `handoff-resumer`: Resume from handoff
-- Output: `thoughts/shared/handoffs/`
-
 ## Commands
 
 | Command | Description |
 |---------|-------------|
 | `/init` | Initialize project with ARCHITECTURE.md and CODE_STYLE.md |
 | `/ledger` | Create or update continuity ledger for session state |
-| `/search` | Search past handoffs, plans, and ledgers |
+| `/search` | Search past plans and ledgers |
 
 ## Agents
 
@@ -207,8 +224,6 @@ Save/resume session state for continuity:
 | reviewer | subagent | claude-opus-4-5 | Review correctness and style |
 | ledger-creator | subagent | claude-sonnet | Create/update continuity ledgers |
 | artifact-searcher | subagent | claude-sonnet | Search past work for precedent |
-| handoff-creator | subagent | claude-opus-4-5 | Save session state |
-| handoff-resumer | subagent | claude-opus-4-5 | Resume from handoff |
 
 ## Tools
 
@@ -217,7 +232,7 @@ Save/resume session state for continuity:
 | `ast_grep_search` | AST-aware code pattern search |
 | `ast_grep_replace` | AST-aware code pattern replacement |
 | `look_at` | Extract file structure for large files |
-| `artifact_search` | Search past handoffs, plans, and ledgers |
+| `artifact_search` | Search past plans and ledgers |
 | `background_task` | Run long-running tasks in background |
 | `background_output` | Check background task status/output |
 | `background_cancel` | Cancel background tasks |
@@ -229,7 +244,8 @@ Save/resume session state for continuity:
 |------|-------------|
 | Think Mode | Keywords like "think hard" enable 32k token thinking budget |
 | Ledger Loader | Injects continuity ledger into system prompt |
-| Auto-Clear Ledger | At 80% context, saves ledger + handoff and clears session |
+| Auto-Clear Ledger | At 80% context, saves ledger with file ops and clears session |
+| File Ops Tracker | Tracks read/write/edit tool calls for deterministic file operation logging |
 | Artifact Auto-Index | Indexes artifacts when written to thoughts/ directories |
 | Auto-Compact | Summarizes session when hitting token limits |
 | Context Injector | Injects ARCHITECTURE.md, CODE_STYLE.md, .cursorrules |
@@ -276,8 +292,7 @@ micode/
     ├── ledgers/        # Continuity ledgers
     └── shared/
         ├── designs/    # Brainstorm outputs
-        ├── plans/      # Implementation plans
-        └── handoffs/   # Session handoffs
+        └── plans/      # Implementation plans
 ```
 
 ## Development
@@ -341,3 +356,4 @@ Built on techniques from:
 
 - **[oh-my-opencode](https://github.com/code-yeongyu/oh-my-opencode)** - OpenCode plugin architecture, agent orchestration patterns, and trusted publishing setup
 - **[HumanLayer ACE-FCA](https://github.com/humanlayer/12-factor-agents)** - Advanced Context Engineering for Coding Agents, structured workflows, and the research → plan → implement methodology
+- **[Factory.ai Context Compression](https://factory.ai/blog/context-compression)** - Structured compaction research showing that anchored iterative summarization with deterministic file tracking outperforms generic compression

src/agents/commander.ts

@@ -76,9 +76,9 @@ Just do it - including obvious follow-up actions.
 <rule>Reference plan file in commit body</rule>
 </phase>
 
-<phase name="handoff">
-<agent name="handoff-creator">Save session state</agent>
-<agent name="handoff-resumer">Resume from handoff</agent>
+<phase name="ledger" trigger="context getting full or session ending">
+<action>System auto-updates ledger at 80% context usage</action>
+<output>thoughts/ledgers/CONTINUITY_{session-name}.md</output>
 </phase>
 </workflow>
 
@@ -89,8 +89,7 @@ Just do it - including obvious follow-up actions.
 <agent name="pattern-finder" mode="subagent" purpose="Find existing patterns"/>
 <agent name="planner" mode="subagent" purpose="Create detailed implementation plans"/>
 <agent name="executor" mode="subagent" purpose="Execute plan (runs implementer then reviewer automatically)"/>
-<agent name="handoff-creator" mode="subagent" purpose="Create handoff docs"/>
-<agent name="handoff-resumer" mode="subagent" purpose="Resume from handoffs"/>
+<agent name="ledger-creator" mode="subagent" purpose="Create/update continuity ledgers"/>
 <parallelization>
 <safe>locator, analyzer, pattern-finder</safe>
 <sequential>planner then executor</sequential>

src/agents/index.ts

@@ -7,8 +7,6 @@ import { plannerAgent } from "./planner";
 import { implementerAgent } from "./implementer";
 import { reviewerAgent } from "./reviewer";
 import { executorAgent } from "./executor";
-import { handoffCreatorAgent } from "./handoff-creator";
-import { handoffResumerAgent } from "./handoff-resumer";
 import { primaryAgent, PRIMARY_AGENT_NAME } from "./commander";
 import { projectInitializerAgent } from "./project-initializer";
 import { ledgerCreatorAgent } from "./ledger-creator";
@@ -24,8 +22,6 @@ export const agents: Record<string, AgentConfig> = {
   implementer: implementerAgent,
   reviewer: reviewerAgent,
   executor: executorAgent,
-  "handoff-creator": handoffCreatorAgent,
-  "handoff-resumer": handoffResumerAgent,
   "project-initializer": projectInitializerAgent,
   "ledger-creator": ledgerCreatorAgent,
   "artifact-searcher": artifactSearcherAgent,
@@ -42,8 +38,6 @@ export {
   implementerAgent,
   reviewerAgent,
   executorAgent,
-  handoffCreatorAgent,
-  handoffResumerAgent,
   projectInitializerAgent,
   ledgerCreatorAgent,
   artifactSearcherAgent,