chore: remove progress markdown files (track via Forgejo issues only)

This commit is contained in:
2026-09-15 00:07:04 +09:00
parent 379aa5ce4d
commit e62860d232
3 changed files with 0 additions and 671 deletions
-263
View File
@@ -1,263 +0,0 @@
# CRITICAL FIXES NEEDED - Poimen Memory Service
## STATUS: Service Non-Functional ❌
**Root Issues Blocking Service**:
1. ✅ HTTP handler deadlock fixed (schema init error handling)
2. ❌ Server initialization hangs during schema or startup (logs stop after `l2_l1_edges`)
3. ❌ Ingest pipeline NOT implemented (just raw vector storage, no entities/edges)
4. ❌ Temporal schema missing (no t_valid, t_invalid, version tracking)
5. ❌ GRM gate not integrated (no memorability scores, confidence)
6. ❌ Query doesn't use knowledge graph (just vector search)
7. ❌ Compaction disabled
8. ❌ Verification gates missing
---
## STEP 1: Fix Server Startup Hang ⚠️
**Current Issue**: Server hangs during initialization after schema creation.
**Suspected causes**:
- OptimizerServiceBuilder.build() getting stuck
- AccessGuard creation blocking
- Background task spawning deadlock
**Fix**:
```rust
// In http_server.rs:316-325
// Wrap in timeout or disable non-essentials
let optimizer_service = match tokio::time::timeout(
Duration::from_secs(5),
async { mem_core::optimizer::OptimizerServiceBuilder::new().build() }
).await {
Ok(Ok(service)) => Some(Arc::new(service)),
_ => {
tracing::warn!("Optimizer initialization skipped (timeout or error)");
None
}
};
```
**Test**: `./target/release/mem serve --port 9999` should reach "Starting HTTP server" within 10s
---
## STEP 2: Implement Ingest Pipeline (HIGH PRIORITY)
**Current Implementation** (`ingest_worker.rs`):
```rust
// Just stores raw chunks + embeddings
store_chunk_l0(&l0_chunk)
store_memory_l1(&l1_memory, &embedding)
```
**Expected Implementation**:
```rust
// 1. Extract entities (entity_extractor)
let entities = entity_extractor.extract(&content).await?;
// 2. Extract facts + edges (fact_extractor)
let facts = fact_extractor.extract(&content, entities).await?;
// 3. Create temporal edges with GRM gate
for fact in facts {
let edge = TemporalEdge {
source: fact.source_entity,
target: fact.target_entity,
relation: fact.relation,
fact: fact.text,
t_valid: now(),
t_invalid: None,
confidence: grm_gate.score(&fact)?, // ← GRM gate
version: 1,
};
edge_repo.insert(&edge).await?;
}
// 4. Check contradictions + queue for review
for edge in edges {
if contradiction_detector.detect(&edge, existing_edges)? {
review_queue.enqueue(&edge).await?;
}
}
```
**Files to modify**:
- `crates/mem-cli/src/ingest_worker.rs` (core ingest logic)
- `crates/mem-ingest/src/ingest_pipeline.rs` (entity + fact extraction)
- `crates/mem-ingest/src/contradiction_detector.rs` (pre-filter + review)
---
## STEP 3: Update Storage Schema (MEDIUM PRIORITY)
**Missing fields**:
```sql
ALTER TABLE memories_l1 ADD COLUMN (
t_valid TIMESTAMP NOT NULL DEFAULT NOW(),
t_invalid TIMESTAMP,
confidence FLOAT DEFAULT 0.5,
version INT DEFAULT 1,
memorability_score INT,
contribution_date TIMESTAMP
);
ALTER TABLE l1_l0_edges MODIFY TO (
l1_id UUID,
l0_id UUID,
relation_type VARCHAR,
fact TEXT,
t_valid TIMESTAMP DEFAULT NOW(),
t_invalid TIMESTAMP,
confidence FLOAT,
contradiction_flag BOOL DEFAULT FALSE,
review_queue_id UUID,
version INT DEFAULT 1,
PRIMARY KEY (l1_id, l0_id, version)
);
```
**Migration script**: `crates/mem-store/migrations/003_temporal_grm_schema.sql`
---
## STEP 4: Wire Query Handler to Knowledge Graph (MEDIUM PRIORITY)
**Current** (`query_handler` in http_server.rs):
```rust
async fn query_handler(...) -> HttpResponse {
// Just semantic search
let results = vector_search(query)?;
HttpResponse::Ok().json(results)
}
```
**Expected**:
```rust
async fn query_handler(query: QueryRequest) -> HttpResponse {
// 1. Semantic search on embeddings
let initial_results = vector_search(&query.text)?;
// 2. Follow edges (graph traversal)
let mut expanded = vec![];
for result in initial_results {
expanded.push(result);
// Get related entities via edges
let related = edge_repo.find_by_source(&result.entity_id).await?;
expanded.extend(related);
}
// 3. Apply temporal filters
expanded.retain(|e| e.t_valid <= now() && (e.t_invalid.is_none() || e.t_invalid > now()));
// 4. Sort by confidence + recency
expanded.sort_by(|a, b| {
b.confidence.partial_cmp(&a.confidence)
.then_with(|| b.t_valid.cmp(&a.t_valid))
});
// 5. Apply compaction/cache alignment
for item in &mut expanded {
item.text = optimizer.compress(item.text)?;
}
HttpResponse::Ok().json(MemoryResponse {
entities: expanded,
confidence_scores: compute_scores(&expanded),
})
}
```
---
## STEP 5: Enable Compaction Endpoint (LOW PRIORITY)
**Current**: Code exists but never called.
**Fix**: Add K8s CronJob that calls `POST /memory/compact` daily:
```yaml
apiVersion: batch/v1
kind: CronJob
metadata:
name: memory-compaction
spec:
schedule: "0 2 * * *" # 2 AM UTC
jobTemplate:
spec:
template:
spec:
containers:
- name: compact
image: bitnami/curl:latest
command:
- curl
- -X POST
- -H "Authorization: Bearer $ADMIN_TOKEN"
- http://poimen-memory:8080/memory/compact
restartPolicy: OnFailure
```
---
## STEP 6: Add Verification Gates (LOW PRIORITY)
**Missing**: `GET /memory/verify` endpoint that checks M1.8, M2.8, M3.7, M8.9 gates
---
## IMPLEMENTATION ORDER
1. **FIX STARTUP** (1 hour) → Get server running
2. **INGEST PIPELINE** (3 hours) → Wire entity + fact extraction
3. **TEMPORAL SCHEMA** (1 hour) → Add missing columns
4. **QUERY HANDLER** (2 hours) → Implement graph traversal
5. **COMPACTION** (1 hour) → Add CronJob
6. **GATES** (2 hours) → Quality verification
**Total**: ~10 hours to full working system
---
## TEST PLAN
```bash
# 1. Server starts
curl http://localhost:9999/health
# Expected: {"status":"ok","uptime_seconds":N}
# 2. Ingest works
curl -X POST http://localhost:9999/memory/ingest \
-H "Content-Type: application/json" \
-d '{"project":"test","source":"test://1","ingest_id":"i1","records":[{"role":"user","text":"Hello world","timestamp":"2026-01-08T16:00:00Z","source_position":0}]}'
# Expected: {"ingest_id":"i1","status":"pending",...}
# 3. Query returns entities with edges
curl -X POST http://localhost:9999/memory/query \
-H "Content-Type: application/json" \
-d '{"project":"test","query":"hello"}'
# Expected: {"results":[{"type":"entity","name":"...","edges":[...]}]}
# 4. Temporal filtering works
curl http://localhost:9999/memory/query?project=test&temporal_floor=2026-01-01
# 5. Compaction works
curl -X POST http://localhost:9999/memory/compact
# Expected: {"phase":"completed","records_deduplicated":N}
```
---
## FILES MODIFIED SO FAR
`crates/mem-cli/src/http_server.rs` - Added error handling for schema init
---
## NEXT SESSION TODO
- [ ] Fix server startup hang (debug OptimizerService)
- [ ] Implement ingest_worker to call entity_extractor + fact_extractor
- [ ] Add temporal columns to schema
- [ ] Update query_handler to traverse edges
- [ ] Test end-to-end with sample data
-217
View File
@@ -1,217 +0,0 @@
# Monitoring Agent: Implementation Tasks
**Milestone**: `monitoring-agent`
**Status**: 🔧 Not started
**Duration**: 4-6 weeks
**Effort**: ~1,500 LOC
---
## Phase 1: Temporal Setup (3-5 days)
### Task 1.1: Deploy Temporal Server in K8s
- [ ] StatefulSet configuration (persistence)
- [ ] PostgreSQL event log backend
- [ ] ElasticSearch for visibility
- [ ] K8s manifests in `k8s/temporal/`
- [ ] Health checks + readiness probes
- **Effort**: 150 LOC | **Time**: 2 days
- **Dependencies**: None
- **Blocks**: Phase 2
### Task 1.2: Add Temporal SDK to Rust Project
- [ ] Add `temporal-rust-sdk` to `Cargo.toml`
- [ ] Create `crates/mem-temporal/` workspace crate
- [ ] Worker registration + gRPC connection
- [ ] Activity executor setup
- [ ] Workflow executor setup
- **Effort**: 200 LOC | **Time**: 1 day
- **Dependencies**: 1.1
- **Blocks**: Phase 2
### Task 1.3: Temporal Configuration + Secrets
- [ ] Environment variables (TEMPORAL_HOST, TEMPORAL_NAMESPACE)
- [ ] Worker identity configuration
- [ ] Task queue setup (synthesis-queue, compaction-queue)
- **Effort**: 50 LOC | **Time**: 4 hours
- **Dependencies**: 1.1, 1.2
- **Blocks**: Phase 2
---
## Phase 2: Agent Workflows (1-2 weeks)
### Task 2.1: Synthesis Workflow Definition
- [ ] `crates/mem-temporal/src/workflows/synthesis_workflow.rs`
- [ ] Workflow orchestration logic
- [ ] Activity composition (health check → synthesis → logging → metrics)
- [ ] Retry policies (exponential backoff, max 5 retries)
- [ ] Heartbeat configuration (every 10s)
- **Effort**: 200 LOC | **Time**: 3 days
- **Dependencies**: 1.2, 1.3
- **Blocks**: 2.3, 2.4
### Task 2.2: Synthesis Activities (5 activities)
- [ ] `MonitorMemoryHealth` activity
- GET /health check
- Latency measurement
- Failure detection
- [ ] `ExecuteSynthesis` activity
- POST /memory/synthesize call
- LLM integration
- Heartbeat emission
- [ ] `LogSynthesisResult` activity
- POST /memory/ingest (audit)
- Temporal audit trail
- [ ] `UpdateCacheMetrics` activity
- Metric recording
- Performance tracking
- [ ] `CoordinateCompaction` activity
- Signal to compaction agent
- Readiness check
- **Effort**: 250 LOC | **Time**: 4 days
- **Dependencies**: 2.1
- **Blocks**: 2.3
### Task 2.3: Compaction Workflow Definition
- [ ] `crates/mem-temporal/src/workflows/compaction_workflow.rs`
- [ ] 4-stage orchestration (identify → dedup → gc → invalidate)
- [ ] Failure handling + rollback strategy
- **Effort**: 150 LOC | **Time**: 2 days
- **Dependencies**: 1.2, 1.3
- **Blocks**: 2.4
### Task 2.4: Compaction Activities (4 activities)
- [ ] `IdentifyDuplicates` activity
- [ ] `DeduplicateEdges` activity
- [ ] `GarbageCollection` activity
- [ ] `InvalidateCache` activity
- **Effort**: 200 LOC | **Time**: 3 days
- **Dependencies**: 2.3
- **Blocks**: Integration tests
### Task 2.5: Worker + Task Queue Registration
- [ ] Activity worker setup
- [ ] Workflow worker setup
- [ ] Task queue polling
- [ ] Namespace configuration
- **Effort**: 100 LOC | **Time**: 1 day
- **Dependencies**: 2.1-2.4
- **Blocks**: Phase 3
---
## Phase 3: Agent Self-Awareness (2-3 weeks)
### Task 3.1: AGENT_PROMPT Entity Type
- [ ] Schema: New entity type in memory_entity
- [ ] Repository: `synthesis_cache_repo.rs` (get_agent_prompt)
- [ ] Migration: Add to entity type enum
- [ ] Activity: Load prompt on agent startup
- **Effort**: 100 LOC | **Time**: 1 day
- **Dependencies**: Memory service
- **Blocks**: 3.2
### Task 3.2: AGENT_SKILL Linking
- [ ] Edge type: agent → skill relationships
- [ ] Repository methods: link_agent_to_skill, get_agent_skills
- [ ] Confidence tracking per skill
- [ ] Success rate calculation
- **Effort**: 80 LOC | **Time**: 1 day
- **Dependencies**: 3.1
- **Blocks**: 3.4
### Task 3.3: AGENT_PERFORMANCE Metrics
- [ ] Entity type: Temporal metrics
- [ ] Repository: Store + query metrics
- [ ] Activity: Log performance data post-execution
- [ ] Time window filtering (last_7_days, last_30_days)
- **Effort**: 120 LOC | **Time**: 2 days
- **Dependencies**: 3.1
- **Blocks**: 3.4
### Task 3.4: Agent Decision Tracking + Learning
- [ ] Edge type: agent_decision_outcome
- [ ] Decision logging (parameter, value, confidence before)
- [ ] Outcome recording (result, metric)
- [ ] Confidence evolution (update after outcome)
- [ ] Learning loop in agent code
- **Effort**: 200 LOC | **Time**: 3 days
- **Dependencies**: 3.1-3.3
- **Blocks**: 3.5
### Task 3.5: Agent Audit Trail Integration
- [ ] Dual audit: Temporal history + Memory entities
- [ ] Query interface for reviewers
- [ ] Temporal CLI integration
- [ ] Retention policy (365 days)
- **Effort**: 100 LOC | **Time**: 1 day
- **Dependencies**: 3.1-3.4
- **Blocks**: Testing
---
## Testing & Documentation
### Task 4.1: Integration Tests
- [ ] Workflow execution end-to-end
- [ ] Activity retry behavior
- [ ] Heartbeat detection
- [ ] Failure recovery
- [ ] State replay on restart
- **Effort**: 300 LOC | **Time**: 3 days
- **Dependencies**: Phase 2 complete
- **Blocks**: Integration
### Task 4.2: Monitoring & Observability
- [ ] Temporal UI setup (temporal.riotpiao.com)
- [ ] Prometheus metrics export
- [ ] Alerting rules (workflow timeout, activity failure)
- [ ] Grafana dashboards
- **Effort**: 150 LOC | **Time**: 2 days
- **Dependencies**: Phase 1 complete
- **Blocks**: Production
### Task 4.3: Documentation
- [ ] Agent architecture diagram
- [ ] Workflow execution flow
- [ ] Operational runbook
- [ ] Troubleshooting guide
- **Effort**: 50 LOC | **Time**: 1 day
- **Dependencies**: All phases
- **Blocks**: Release
---
## Credentials Status
**SOPS Encrypted**: `k8s/app/memory-agent-secrets.enc.yaml`
- CLIENT_ID: `memory-agent`
- CLIENT_SECRET: Encrypted
- TOKEN_URL: `https://authentik.riotpiao.com/application/o/token/`
- AUTHENTIK_ISSUER: `https://authentik.riotpiao.com/application/o/memory-agent/`
**JWT Auth Verified**: `memory-agent` credentials working
- Test result: Token obtained successfully
- Expiry: 1 hour (3600s)
- Scopes: Default (sufficient for LLM operations)
---
## Timeline
```
Week 1 (Phase 1): Temporal setup
Week 2-3 (Phase 2): Agent workflows
Week 4-5 (Phase 3): Self-awareness
Week 6 (Testing + Docs): Integration + release
```
**Start Date**: TBD
**Target End Date**: TBD (+4-6 weeks)
-191
View File
@@ -1,191 +0,0 @@
# Current Status - Poimen Memory Service (2026-01-08)
## ✅ COMPLETED THIS SESSION
### 1. Removed AccessGuard RBAC (Blocker Issue #1)
-~~AccessGuard initialization~~ REMOVED
-~~RBAC checks in handlers~~ REMOVED
-~~Permission-based access control~~ DEFERRED
- ✅ Code now compiles with `cargo build --release`
- ✅ Binary created: `target/release/mem`
### 2. HTTP Handler Initialization Fixed
- ✅ Added error handling for schema initialization
- ✅ Server reaches "Starting HTTP server" log message
- ✅ HTTP server binds to port (processes created)
## ⚠️ CURRENT ISSUE
**Server binds to port but exits immediately (silent failure)**
Process is created and runs `serve` command, but:
- Process exits with code 0 (clean exit, no crash)
- No HTTP requests answered (port refuses connections)
- Logs don't show "listening on 0.0.0.0:8080" message
**Suspected cause**: Something in the handler initialization or routing setup is blocking/panicking but not showing in logs.
## 🔧 DEBUGGING STEPS NEEDED
1. Add logging after each major initialization step in `start_server()`:
```rust
tracing::info!("About to create AppState");
let state = web::Data::new(AppState { ... });
tracing::info!("AppState created");
tracing::info!("About to create HttpServer");
HttpServer::new(move || { ... })
tracing::info!("HttpServer created, about to bind");
.bind(("0.0.0.0", port))?
tracing::info!("Bound to port {}", port);
.run()
tracing::info!("About to run()");
.await?;
tracing::info!("Server running");
```
2. Run with `RUST_BACKTRACE=1` to see panics
3. Check if the issue is in handler route registration
## 📋 NEXT PRIORITY FIXES (AFTER SERVER RUNS)
### Phase 1: INGEST PIPELINE ⭐ CRITICAL
**File**: `crates/mem-cli/src/ingest_worker.rs`
Currently: Just stores raw vectors
```rust
// WRONG - just vector storage
store_chunk_l0(&l0_chunk);
store_memory_l1(&l1_memory);
```
Should: Extract entities + facts + edges
```rust
// 1. Extract entities
let entities = entity_extractor.extract(&content).await?;
// 2. Extract facts/relationships
let facts = fact_extractor.extract(&content, &entities).await?;
// 3. Create temporal edges
for fact in facts {
let edge = TemporalEdge {
source: fact.source_entity,
target: fact.target_entity,
relation: fact.relation,
fact: fact.text,
t_valid: now(),
t_invalid: None,
confidence: 0.8, // GRM gate score
version: 1,
};
edge_repo.insert(&edge).await?;
}
// 4. Queue contradictions for review
for edge in &edges {
if contradiction_detector.detect(edge, existing_edges)? {
review_queue.enqueue(edge).await?;
}
}
```
### Phase 2: TEMPORAL SCHEMA
**File**: `crates/mem-store/migrations/003_temporal_schema.sql`
Add columns:
- `t_valid TIMESTAMP NOT NULL DEFAULT NOW()`
- `t_invalid TIMESTAMP`
- `confidence FLOAT DEFAULT 0.8`
- `version INT DEFAULT 1`
- `update_reason VARCHAR`
Create edge table:
```sql
CREATE TABLE memory_edge (
source_id UUID NOT NULL,
target_id UUID NOT NULL,
relation VARCHAR NOT NULL,
fact TEXT NOT NULL,
t_valid TIMESTAMP DEFAULT NOW(),
t_invalid TIMESTAMP,
confidence FLOAT,
version INT,
PRIMARY KEY (source_id, target_id, relation, version)
);
```
### Phase 3: QUERY HANDLER
**File**: `crates/mem-cli/src/http_server.rs`
Change `query_handler()` from vector-only to graph-aware:
```rust
// 1. Vector search
let results = semantic_search(query)?;
// 2. Follow edges
let mut expanded = results;
for entity in results {
let related = edge_repo.find_by_source(&entity.id).await?;
expanded.extend(related);
}
// 3. Apply temporal filter
expanded.retain(|e| is_valid_at_time(e, now()));
// 4. Sort by confidence + recency
expanded.sort_by_key(|e| (-e.confidence, -e.t_valid));
// 5. Return
HttpResponse::Ok().json(expanded)
```
### Phase 4: END-TO-END TESTING
```bash
# 1. Ingest with entities + facts
POST /memory/ingest
{
"project": "test",
"source": "transcript://session-1",
"ingest_id": "i-001",
"records": [{"role": "user", "text": "Kubernetes port conflict...", ...}]
}
# Expected: {"ingest_id":"i-001","status":"pending"}
# 2. Check ingest status
GET /memory/ingest/i-001
# Expected: {"status":"done","entities_count":5,"edges_count":3}
# 3. Query returns graph
POST /memory/query
{"project":"test","query":"port conflict resolution"}
# Expected: {"results":[
# {"type":"entity","name":"Kubernetes","edges":[...]},
# {"type":"entity","name":"Port","edges":[...]},
# {"type":"fact","source":"Kubernetes","target":"Port","relation":"has-conflict"}
# ]}
```
## FILES MODIFIED
✅ `crates/mem-cli/src/http_server.rs` - Removed RBAC, added error handling
✅ Created `STATUS_CURRENT.md` - This file
## TIMELINE
- **2026-01-08 16:00**: Fixed HTTP handlers, removed RBAC blocker
- **2026-01-08 16:30**: Server init working, but exits on startup
- **2026-01-08 16:40**: Debugging server binding issue
## KEY DECISIONS
1. **RBAC deferred**: MVP focuses on core ingest/query, auth added later
2. **Temporal-first**: All edges must have t_valid/t_invalid for graph compaction
3. **GRM gate integrated at ingest time**: Confidence scores assigned when facts extracted
4. **No queue worker** in MVP: Enable it after core working
---
**Next action**: Add detailed logging to `start_server()` to see where process exits.