Phase 4: Confidence Scoring & Tooling (Memory Enhancement Layer)

[rjmurillo-bot]

Phase 4: Confidence Scoring & Tooling

Epic: #990 - Memory Enhancement Layer for Serena + Forgetful
PRD: PRD-memory-enhancement-layer-for-serena-forgetful.md
Duration: 1 week
Priority: P2

Deliverables

• Confidence calculation based on verification history
• \serena.py\ - helper to add citations to existing memories
• Integration with
  eflect\ skill for auto-citations
• Claude Code skill wrapper (\SKILL.md)

Exit Criteria

• Memories track confidence over time
• Easy to add/update citations via CLI or skill

Dependencies

• Phase 1: Citation Schema & Verification (Phase 1: Citation Schema & Verification (Memory Enhancement Layer) #997) - requires models.py and citations.py
• Phase 2: Graph Traversal (Phase 2: Graph Traversal (Memory Enhancement Layer) #998) - for confidence propagation across links
• Phase 3: Health Reporting (Phase 3: Health Reporting & CI Integration (Memory Enhancement Layer) #999) - for verification history tracking

[traycerai]

Plan

Observations

Phase 4 builds on the memory enhancement layer foundation (Phases 1-3) to add confidence scoring and developer-friendly tooling. The existing codebase uses PowerShell for skills (file:.claude/skills/github/SKILL.md, file:.claude/skills/codeql-scan/SKILL.md) and has a Python project structure (file:pyproject.toml) ready for the memory_enhancement package. The reflect skill (file:.claude/skills/reflect/SKILL.md) already writes to Serena memories and needs integration with citation tracking. The PRD (file:.agents/specs/PRD-memory-enhancement-layer-for-serena-forgetful.md) specifies YAML frontmatter for citations, confidence calculation based on verification history, and a CLI-first approach with optional skill wrapper.

Approach

The implementation follows a layered approach: (1) extend the Python package with confidence scoring logic in serena.py, (2) create a CLI helper for adding citations to existing memories, (3) integrate with the reflect skill to auto-capture citations when learnings reference code, and (4) wrap everything in a Claude Code skill for discoverability. This approach maintains backward compatibility with existing Serena memories (citations are optional), leverages the established skill pattern (SKILL.md + scripts), and provides both programmatic (Python API) and interactive (CLI + skill) interfaces. The confidence score is calculated as valid_citations / total_citations and stored in YAML frontmatter alongside last_verified timestamp.

Implementation Steps

1. Extend Python Package with Confidence Scoring

File: src/memory_enhancement/serena.py

Implement confidence calculation logic that:

• Reads verification history from last_verified timestamp in YAML frontmatter
• Calculates confidence as ratio of valid citations to total citations
• Updates confidence score after each verification run
• Provides update_confidence() function that takes a Memory object and VerificationResult, updates the frontmatter fields (confidence, last_verified), and writes back to disk
• Implements add_citation_to_memory() as specified in PRD (lines 773-808) to append citations to existing memories via YAML frontmatter manipulation using python-frontmatter library

Dependencies:

• Phase 1: models.py (Memory, Citation dataclasses)
• Phase 1: citations.py (verify_memory, VerificationResult)
• Add python-frontmatter to pyproject.toml dependencies

Key Functions:

• update_confidence(memory: Memory, verification: VerificationResult) -> None - Updates confidence and last_verified in frontmatter
• add_citation_to_memory(memory_path: Path, file_path: str, line: int, snippet: str) -> None - Adds citation to existing memory (already in PRD)
• get_confidence_history(memory: Memory) -> list[tuple[datetime, float]] - Retrieves historical confidence scores if tracked

2. Create CLI Helper for Citation Management

File: src/memory_enhancement/__main__.py

Extend the CLI entrypoint (already specified in PRD lines 813-916) with citation management commands:

New Subcommands:

• add-citation <memory-id> --file <path> --line <num> --snippet <text> - Add citation to existing memory
• update-confidence <memory-id> - Recalculate and update confidence score
• list-citations <memory-id> - Display all citations for a memory with verification status

Implementation:

• Parse arguments using argparse (consistent with existing commands)
• Call add_citation_to_memory() from serena.py
• Display user-friendly output with emoji indicators (✅ valid, ❌ stale)
• Support --json flag for programmatic usage
• Add --dry-run flag to preview changes without writing

Exit Codes (following ADR-035 pattern from codeql-scan):

• 0: Success
• 1: Validation failed (stale citations)
• 2: Invalid arguments
• 3: File I/O error

3. Integrate with Reflect Skill for Auto-Citations

File: .claude/skills/reflect/SKILL.md

Update the reflect skill to automatically capture citations when learnings reference code:

Phase 4 Enhancement (add after line 297):

When persisting learnings that reference specific code locations:

1. Detect code references in learning text (e.g., "in src/client/constants.ts line 42")
2. Extract citation metadata: file path, line number, snippet
3. Add citations to frontmatter using python -m memory_enhancement add-citation
4. Update confidence score based on initial verification

Detection Patterns:

• Inline code references: `path/to/file.ext` line N
• Function references: `functionName()` in `file.ext`
• Explicit citations: "See: file.ext:42"

Integration Point:
After user approves learnings (Phase 4, line 318), before writing to Serena:

1. Extract code references from learning text
2. For each reference:
   - Parse file path, line number, snippet
   - Call: python -m memory_enhancement add-citation <memory-id> --file <path> --line <num> --snippet <text>
3. Proceed with normal Serena MCP write

Fallback: If citation extraction fails, proceed without citations (non-blocking)

4. Create Claude Code Skill Wrapper

File: .claude/skills/memory-enhancement/SKILL.md

Create a new skill following the established pattern (file:.claude/skills/SkillForge/assets/templates/skill-md-template.md):

Frontmatter:

---
name: memory-enhancement
version: 1.0.0
description: Manage memory citations, verify code references, and track confidence scores. Use when adding citations to memories, checking memory health, or verifying code references are still valid.
license: MIT
model: claude-sonnet-4-5
metadata:
  domains: [memory, citations, verification]
  type: utility
  adr: ADR-007, ADR-037
---

Triggers:

• "add citation to memory"
• "verify memory citations"
• "check memory health"
• "update memory confidence"

Decision Tree:

Need memory enhancement?
├─ Add citation to memory → add-citation command
├─ Verify citations → verify or verify-all command
├─ Check memory health → health command
├─ Traverse memory graph → graph command
└─ Update confidence → update-confidence command

Process Sections:

• Phase 1: Identify Target Memory - Locate memory file by ID or path
• Phase 2: Add/Verify Citations - Use CLI commands with structured output
• Phase 3: Update Confidence - Recalculate based on verification results
• Phase 4: Report Results - Display summary with actionable recommendations

Script Reference Table:

Operation	CLI Command	Key Parameters
Add citation	python -m memory_enhancement add-citation	<memory-id>, --file, --line, --snippet
Verify memory	python -m memory_enhancement verify	<memory-id>, --json
Verify all	python -m memory_enhancement verify-all	--dir, --json
Health report	python -m memory_enhancement health	--format (markdown/json)
Update confidence	python -m memory_enhancement update-confidence	<memory-id>

Anti-Patterns:

• Don't add citations without verifying file exists
• Don't skip confidence updates after verification
• Don't use absolute paths (use repo-relative paths)
• Don't add duplicate citations (check existing first)

Integration with Existing Skills:

• reflect - Auto-capture citations from learnings
• memory - Verify citations during memory search
• curating-memories - Update citations when memories change

5. Add Python Dependencies

File: pyproject.toml

Add required dependencies for Phase 4:

dependencies = [
    "anthropic>=0.39.0",
    "python-frontmatter>=1.0.0",  # YAML frontmatter parsing
]

6. Create Unit Tests

File: tests/test_serena.py

Test confidence scoring and citation management:

• test_update_confidence_with_valid_citations() - Confidence = 1.0 when all valid
• test_update_confidence_with_stale_citations() - Confidence < 1.0 when some stale
• test_add_citation_to_memory() - Citation added to frontmatter
• test_add_citation_preserves_existing() - Existing citations not lost
• test_confidence_without_citations() - Uses default confidence (0.5)

File: tests/test_cli_citations.py

Test CLI citation commands:

• test_add_citation_command() - CLI adds citation successfully
• test_add_citation_invalid_file() - Error handling for missing files
• test_update_confidence_command() - CLI updates confidence
• test_list_citations_command() - CLI displays citations with status

7. Update Documentation

File: .claude/skills/memory-enhancement/references/examples.md

Create examples document with:

• Adding citation to existing memory
• Verifying citations in batch
• Integrating with reflect skill workflow
• Interpreting confidence scores
• Troubleshooting common issues

File: .claude/skills/memory-enhancement/references/confidence-scoring.md

Document confidence calculation:

• Formula: valid_citations / total_citations
• Interpretation: 0.0-0.5 (low), 0.5-0.8 (medium), 0.8-1.0 (high)
• When to update: After code changes, during health checks, on-demand
• Historical tracking: Store verification timestamps in frontmatter

File: AGENTS.md

Update memory system section (lines 958-1133) to reference new skill:

• Add memory-enhancement to skill catalog
• Document citation workflow
• Link to confidence scoring guide

8. Integration Testing

Workflow:

1. Create test memory in .serena/memories/test-memory.md
2. Add citation: python -m memory_enhancement add-citation test-memory --file src/test.py --line 10 --snippet "test"
3. Verify citation: python -m memory_enhancement verify test-memory
4. Check confidence: Verify frontmatter contains confidence: 1.0
5. Modify code to invalidate citation
6. Re-verify: python -m memory_enhancement verify test-memory
7. Check confidence: Verify frontmatter contains confidence: 0.0
8. Test skill wrapper: Invoke via Claude Code skill system
9. Test reflect integration: Create learning with code reference, verify citation auto-added

9. CI Integration (Optional)

File: .github/workflows/memory-validation.yml

Create CI workflow as specified in PRD (lines 974-996):

• Trigger on PR changes to .serena/memories/** or src/**
• Run python -m memory_enhancement verify-all --json
• Post comment with stale memory report
• Initially continue-on-error: true (warning only)
• After adoption, make blocking

Visual Overview

sequenceDiagram
    participant User
    participant ReflectSkill as Reflect Skill
    participant CLI as memory_enhancement CLI
    participant Serena as Serena Memory (.md)
    participant Codebase as Codebase Files

    User->>ReflectSkill: "reflect" (capture learnings)
    ReflectSkill->>ReflectSkill: Analyze conversation for code references
    ReflectSkill->>CLI: add-citation (file, line, snippet)
    CLI->>Codebase: Verify file exists
    Codebase-->>CLI: File content
    CLI->>Serena: Update YAML frontmatter (citations)
    CLI->>CLI: Calculate initial confidence
    CLI->>Serena: Update confidence + last_verified
    CLI-->>ReflectSkill: Citation added
    ReflectSkill->>Serena: Write learning content (MCP)
    ReflectSkill-->>User: Learning persisted with citations

    Note over User,Codebase: Later: Verification workflow
    User->>CLI: verify-all
    CLI->>Serena: Read memories with citations
    CLI->>Codebase: Verify each citation
    Codebase-->>CLI: Validation results
    CLI->>CLI: Recalculate confidence
    CLI->>Serena: Update confidence scores
    CLI-->>User: Health report (stale memories)

Loading

Confidence Score Interpretation

Score Range	Interpretation	Action
0.9 - 1.0	High confidence - all citations valid	Trust memory, use in decisions
0.7 - 0.9	Medium confidence - most citations valid	Review stale citations, consider updating
0.5 - 0.7	Low confidence - many stale citations	Update memory or mark obsolete
0.0 - 0.5	Very low confidence - most citations invalid	Memory likely outdated, needs review
No citations	Default (0.5) - no verification data	Add citations to improve confidence

Import In IDE

🤖 Prompt for AI Agents

## Observations

Phase 4 builds on the memory enhancement layer foundation (Phases 1-3) to add confidence scoring and developer-friendly tooling. The existing codebase uses PowerShell for skills (file:`.claude/skills/github/SKILL.md`, file:`.claude/skills/codeql-scan/SKILL.md`) and has a Python project structure (file:`pyproject.toml`) ready for the `memory_enhancement` package. The `reflect` skill (file:`.claude/skills/reflect/SKILL.md`) already writes to Serena memories and needs integration with citation tracking. The PRD (file:`.agents/specs/PRD-memory-enhancement-layer-for-serena-forgetful.md`) specifies YAML frontmatter for citations, confidence calculation based on verification history, and a CLI-first approach with optional skill wrapper.

## Approach

The implementation follows a layered approach: (1) extend the Python package with confidence scoring logic in `serena.py`, (2) create a CLI helper for adding citations to existing memories, (3) integrate with the `reflect` skill to auto-capture citations when learnings reference code, and (4) wrap everything in a Claude Code skill for discoverability. This approach maintains backward compatibility with existing Serena memories (citations are optional), leverages the established skill pattern (SKILL.md + scripts), and provides both programmatic (Python API) and interactive (CLI + skill) interfaces. The confidence score is calculated as `valid_citations / total_citations` and stored in YAML frontmatter alongside `last_verified` timestamp.

## Implementation Steps

### 1. Extend Python Package with Confidence Scoring

**File**: `src/memory_enhancement/serena.py`

Implement confidence calculation logic that:
- Reads verification history from `last_verified` timestamp in YAML frontmatter
- Calculates confidence as ratio of valid citations to total citations
- Updates confidence score after each verification run
- Provides `update_confidence()` function that takes a `Memory` object and `VerificationResult`, updates the frontmatter fields (`confidence`, `last_verified`), and writes back to disk
- Implements `add_citation_to_memory()` as specified in PRD (lines 773-808) to append citations to existing memories via YAML frontmatter manipulation using `python-frontmatter` library

**Dependencies**:
- Phase 1: `models.py` (Memory, Citation dataclasses)
- Phase 1: `citations.py` (verify_memory, VerificationResult)
- Add `python-frontmatter` to `pyproject.toml` dependencies

**Key Functions**:
- `update_confidence(memory: Memory, verification: VerificationResult) -> None` - Updates confidence and last_verified in frontmatter
- `add_citation_to_memory(memory_path: Path, file_path: str, line: int, snippet: str) -> None` - Adds citation to existing memory (already in PRD)
- `get_confidence_history(memory: Memory) -> list[tuple[datetime, float]]` - Retrieves historical confidence scores if tracked

### 2. Create CLI Helper for Citation Management

**File**: `src/memory_enhancement/__main__.py`

Extend the CLI entrypoint (already specified in PRD lines 813-916) with citation management commands:

**New Subcommands**:
- `add-citation <memory-id> --file <path> --line <num> --snippet <text>` - Add citation to existing memory
- `update-confidence <memory-id>` - Recalculate and update confidence score
- `list-citations <memory-id>` - Display all citations for a memory with verification status

**Implementation**:
- Parse arguments using `argparse` (consistent with existing commands)
- Call `add_citation_to_memory()` from `serena.py`
- Display user-friendly output with emoji indicators (✅ valid, ❌ stale)
- Support `--json` flag for programmatic usage
- Add `--dry-run` flag to preview changes without writing

**Exit Codes** (following ADR-035 pattern from codeql-scan):
- 0: Success
- 1: Validation failed (stale citations)
- 2: Invalid arguments
- 3: File I/O error

### 3. Integrate with Reflect Skill for Auto-Citations

**File**: `.claude/skills/reflect/SKILL.md`

Update the reflect skill to automatically capture citations when learnings reference code:

**Phase 4 Enhancement** (add after line 297):

When persisting learnings that reference specific code locations:
1. **Detect code references** in learning text (e.g., "in `src/client/constants.ts` line 42")
2. **Extract citation metadata**: file path, line number, snippet
3. **Add citations to frontmatter** using `python -m memory_enhancement add-citation`
4. **Update confidence score** based on initial verification

**Detection Patterns**:
- Inline code references: `` `path/to/file.ext` line N ``
- Function references: `` `functionName()` in `file.ext` ``
- Explicit citations: "See: file.ext:42"

**Integration Point**:
After user approves learnings (Phase 4, line 318), before writing to Serena:
```text
1. Extract code references from learning text
2. For each reference:
   - Parse file path, line number, snippet
   - Call: python -m memory_enhancement add-citation <memory-id> --file <path> --line <num> --snippet <text>
3. Proceed with normal Serena MCP write
```

**Fallback**: If citation extraction fails, proceed without citations (non-blocking)

### 4. Create Claude Code Skill Wrapper

**File**: `.claude/skills/memory-enhancement/SKILL.md`

Create a new skill following the established pattern (file:`.claude/skills/SkillForge/assets/templates/skill-md-template.md`):

**Frontmatter**:
```yaml
---
name: memory-enhancement
version: 1.0.0
description: Manage memory citations, verify code references, and track confidence scores. Use when adding citations to memories, checking memory health, or verifying code references are still valid.
license: MIT
model: claude-sonnet-4-5
metadata:
  domains: [memory, citations, verification]
  type: utility
  adr: ADR-007, ADR-037
---
```

**Triggers**:
- "add citation to memory"
- "verify memory citations"
- "check memory health"
- "update memory confidence"

**Decision Tree**:
```text
Need memory enhancement?
├─ Add citation to memory → add-citation command
├─ Verify citations → verify or verify-all command
├─ Check memory health → health command
├─ Traverse memory graph → graph command
└─ Update confidence → update-confidence command
```

**Process Sections**:
- **Phase 1: Identify Target Memory** - Locate memory file by ID or path
- **Phase 2: Add/Verify Citations** - Use CLI commands with structured output
- **Phase 3: Update Confidence** - Recalculate based on verification results
- **Phase 4: Report Results** - Display summary with actionable recommendations

**Script Reference Table**:
| Operation | CLI Command | Key Parameters |
|-----------|-------------|----------------|
| Add citation | `python -m memory_enhancement add-citation` | `<memory-id>`, `--file`, `--line`, `--snippet` |
| Verify memory | `python -m memory_enhancement verify` | `<memory-id>`, `--json` |
| Verify all | `python -m memory_enhancement verify-all` | `--dir`, `--json` |
| Health report | `python -m memory_enhancement health` | `--format` (markdown/json) |
| Update confidence | `python -m memory_enhancement update-confidence` | `<memory-id>` |

**Anti-Patterns**:
- Don't add citations without verifying file exists
- Don't skip confidence updates after verification
- Don't use absolute paths (use repo-relative paths)
- Don't add duplicate citations (check existing first)

**Integration with Existing Skills**:
- `reflect` - Auto-capture citations from learnings
- `memory` - Verify citations during memory search
- `curating-memories` - Update citations when memories change

### 5. Add Python Dependencies

**File**: `pyproject.toml`

Add required dependencies for Phase 4:
```toml
dependencies = [
    "anthropic>=0.39.0",
    "python-frontmatter>=1.0.0",  # YAML frontmatter parsing
]
```

### 6. Create Unit Tests

**File**: `tests/test_serena.py`

Test confidence scoring and citation management:
- `test_update_confidence_with_valid_citations()` - Confidence = 1.0 when all valid
- `test_update_confidence_with_stale_citations()` - Confidence < 1.0 when some stale
- `test_add_citation_to_memory()` - Citation added to frontmatter
- `test_add_citation_preserves_existing()` - Existing citations not lost
- `test_confidence_without_citations()` - Uses default confidence (0.5)

**File**: `tests/test_cli_citations.py`

Test CLI citation commands:
- `test_add_citation_command()` - CLI adds citation successfully
- `test_add_citation_invalid_file()` - Error handling for missing files
- `test_update_confidence_command()` - CLI updates confidence
- `test_list_citations_command()` - CLI displays citations with status

### 7. Update Documentation

**File**: `.claude/skills/memory-enhancement/references/examples.md`

Create examples document with:
- Adding citation to existing memory
- Verifying citations in batch
- Integrating with reflect skill workflow
- Interpreting confidence scores
- Troubleshooting common issues

**File**: `.claude/skills/memory-enhancement/references/confidence-scoring.md`

Document confidence calculation:
- Formula: `valid_citations / total_citations`
- Interpretation: 0.0-0.5 (low), 0.5-0.8 (medium), 0.8-1.0 (high)
- When to update: After code changes, during health checks, on-demand
- Historical tracking: Store verification timestamps in frontmatter

**File**: `AGENTS.md`

Update memory system section (lines 958-1133) to reference new skill:
- Add `memory-enhancement` to skill catalog
- Document citation workflow
- Link to confidence scoring guide

### 8. Integration Testing

**Workflow**:
1. Create test memory in `.serena/memories/test-memory.md`
2. Add citation: `python -m memory_enhancement add-citation test-memory --file src/test.py --line 10 --snippet "test"`
3. Verify citation: `python -m memory_enhancement verify test-memory`
4. Check confidence: Verify frontmatter contains `confidence: 1.0`
5. Modify code to invalidate citation
6. Re-verify: `python -m memory_enhancement verify test-memory`
7. Check confidence: Verify frontmatter contains `confidence: 0.0`
8. Test skill wrapper: Invoke via Claude Code skill system
9. Test reflect integration: Create learning with code reference, verify citation auto-added

### 9. CI Integration (Optional)

**File**: `.github/workflows/memory-validation.yml`

Create CI workflow as specified in PRD (lines 974-996):
- Trigger on PR changes to `.serena/memories/**` or `src/**`
- Run `python -m memory_enhancement verify-all --json`
- Post comment with stale memory report
- Initially `continue-on-error: true` (warning only)
- After adoption, make blocking

## Visual Overview

```mermaid
sequenceDiagram
    participant User
    participant ReflectSkill as Reflect Skill
    participant CLI as memory_enhancement CLI
    participant Serena as Serena Memory (.md)
    participant Codebase as Codebase Files

    User->>ReflectSkill: "reflect" (capture learnings)
    ReflectSkill->>ReflectSkill: Analyze conversation for code references
    ReflectSkill->>CLI: add-citation (file, line, snippet)
    CLI->>Codebase: Verify file exists
    Codebase-->>CLI: File content
    CLI->>Serena: Update YAML frontmatter (citations)
    CLI->>CLI: Calculate initial confidence
    CLI->>Serena: Update confidence + last_verified
    CLI-->>ReflectSkill: Citation added
    ReflectSkill->>Serena: Write learning content (MCP)
    ReflectSkill-->>User: Learning persisted with citations

    Note over User,Codebase: Later: Verification workflow
    User->>CLI: verify-all
    CLI->>Serena: Read memories with citations
    CLI->>Codebase: Verify each citation
    Codebase-->>CLI: Validation results
    CLI->>CLI: Recalculate confidence
    CLI->>Serena: Update confidence scores
    CLI-->>User: Health report (stale memories)
```

## Confidence Score Interpretation

| Score Range | Interpretation | Action |
|-------------|----------------|--------|
| 0.9 - 1.0 | High confidence - all citations valid | Trust memory, use in decisions |
| 0.7 - 0.9 | Medium confidence - most citations valid | Review stale citations, consider updating |
| 0.5 - 0.7 | Low confidence - many stale citations | Update memory or mark obsolete |
| 0.0 - 0.5 | Very low confidence - most citations invalid | Memory likely outdated, needs review |
| No citations | Default (0.5) - no verification data | Add citations to improve confidence |

---

Execution Information

Branch: main
Commit: a729aa5

---

💡 Tips

Supported Commands (Inside Comments)

• Use @traycerai generate to iterate on the previous version of the implementation plan.

Supported Commands (Inside Description)

• Add @traycerai ignore anywhere in the ticket description to prevent this ticket from being processed.
• Add @traycerai branch:<branch-name> anywhere in the ticket description to specify the target branch for the implementation plan.

Community

• Join our Discord Community to get help, request features, and share feedback.
• Follow us on X/Twitter for updates and announcements.

[github-actions]

AI Triage Summary

Note

This issue has been automatically triaged by AI agents

What is AI Triage?

This issue was analyzed by AI agents:

• Analyst Agent: Categorizes the issue and suggests appropriate labels
• Roadmap Agent: Aligns the issue with project milestones and priorities
• Explainer Agent (if escalated): Generates comprehensive PRD

Triage Results

Property	Value
Category	enhancement
Labels	["enhancement","agent-memory","area-infrastructure"]
Priority	P2
Milestone	Not assigned

Categorization Analysis

```json
{
  "labels": ["enhancement", "agent-memory", "area-infrastructure"],
  "category": "enhancement",
  "confidence": 0.95,
  "reasoning": "Structured phase deliverable for memory system enhancement with CLI tooling and skill integration"
}

</details>

<details>
<summary>Roadmap Alignment</summary>

```json
```json
{
  "milestone": "",
  "priority": "P2",
  "epic_alignment": "EPIC-990-memory-enhancement-layer",
  "confidence": 0.95,
  "reasoning": "Phase 4 of defined epic with existing PRD; blocked by Phases 1-3 dependencies",
  "escalate_to_prd": false,
  "escalation_criteria": [],
  "complexity_score": 0
}

</details>

---

<sub>Powered by [AI Issue Triage](https://github.com/rjmurillo/ai-agents) workflow</sub>

[coderabbitai]

📝 CodeRabbit Plan Mode

Generate an implementation plan and prompts that you can use with your favorite coding agent.

• Create Plan

Examples

• Example 1
• Example 2

---

⚠️ Possible Duplicate Issue

• Phase 4: Confidence Scoring #995

🔗 Related PRs

#229 - fix: address PR #226 review comments and labeler workflow issues [merged]
#301 - docs: autonomous PR monitoring prompt and retrospective [merged]
#605 - feat(agents): Phase 1 Spec Layer - spec-generator agent and routing [merged]
#752 - feat(memory): memory system foundation (Session 230) [merged]
#968 - feat(memory): Systematic knowledge bootstrap - 462 learnings migrated to Forgetful [merged]

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!