docs: add visualization backlog tier with dependency tracking

[carlos-alm]

Summary

• Add Tier 1g — Visualization leverage to BACKLOG.md with 12 new items (IDs 59–70):
  • 4 foundation items: scoped export (docs: update performance benchmarks #59), symbol-level edge labels (docs: update performance benchmarks #60), node export annotations (fix: run benchmark after publish to prevent workflow cancellation #61), semantic subgraph labels (docs: reorganize README — lead with problem and value #62)
  • 8 downstream integration items (perf: reduce WASM boundary crossings in JS extractor #63–70) that apply each foundation to specific tools (diff-impact, branch-compare, deps, sequence, audit, context, communities, triage)
• Add Depends on column to every backlog table — makes dependency chains explicit and scannable so completing a foundation item immediately surfaces which follow-on work is unblocked
• Trim verbose cross-tool descriptions from foundation items into dedicated follow-on items, keeping each entry focused on one concern

Test plan

• Markdown renders correctly — all tables have consistent column counts
• Every non-DONE row has a Depends on value (either — or a specific ID)
• Dependency IDs reference valid existing items
• No functional code changes

[claude]

Claude finished @carlos-alm's task —— View job

---

PR Review Completed

Todo List

• Read modified BACKLOG.md file
• Phase 1: Root cause analysis - understand problem being solved
• Phase 2: Configuration change detection (not applicable for docs)
• Phase 3: Test integrity assessment
• Phase 4: Code quality review
• Validate markdown structure and table consistency
• Check dependency reference validity
• Provide final rating and recommendation

Root Cause Analysis

Problem Being Solved: The PR addresses organizational clarity in the backlog by adding dependency tracking and creating a focused visualization tier. The "Depends on" column makes dependency chains explicit and scannable, allowing teams to immediately identify unblocked work when foundation items are completed.

Why This Approach: The approach is methodical and well-structured:

• Foundation items (IDs 59-62) establish core visualization infrastructure
• Downstream items (IDs 63-70) systematically apply each foundation to specific tools
• The dependency column transforms the backlog from flat list to directed graph of work

Risk Assessment: LOW RISK - Pure documentation change with clear organizational benefits.

Validation Results

✅ Markdown Structure

• All tables have consistent 10-column layout (added "Depends on" column)
• Headers properly aligned across all tiers
• Proper markdown escaping and formatting maintained

✅ Dependency Reference Validation

All dependency IDs reference valid existing items:

• ID 56 → "Find similar functions" (foundation for similarity features)
• ID 59 → "Scoped graph rendering" (foundation for focused diagrams)
• ID 60 → "Symbol-level edge labels" (foundation for rich edge annotations)
• ID 61 → "Node export annotations" (foundation for API surface display)
• ID 62 → "Semantic subgraph labels" (foundation for contextual annotations)

✅ Tier Structure Logic

Tier 1g intelligently separates concerns:

• 4 foundation items (59-62): Core visualization infrastructure
• 8 downstream items (63-70): Systematic integration across tools (audit, context, diff-impact, branch-compare, deps, sequence, communities, triage)

This creates a clear implementation path where completing one foundation item immediately unlocks multiple follow-on features.

✅ Content Quality

• Each item has focused scope (trimmed verbose cross-references)
• Problem-fit scores are reasonable (3-5 range for visualization enhancements)
• Benefits clearly articulate user value
• All items marked as zero-dep, foundation-aligned, non-breaking

✅ Consistency Improvements

The "Depends on" column addition is applied systematically:

• DONE items: All show — (no dependencies)
• Tier 1b-1f items: Most show —, appropriate dependencies for embeddings tier (56, 57→56, 58→56)
• Tier 1g items: Proper dependency chains (62→59, 63→59, 64→59, 65→60, 66→60, 67→60, 68→60, 69→61, 70→62)
• Tier 2-3: All show — (standalone features)

Critical Concerns

None identified. This is a well-executed documentation enhancement.

Minor Observations

1. Excellent dependency modeling - The foundation→downstream pattern prevents integration sprawl
2. Scannable prioritization - Teams can now quickly identify which items become unblocked when foundations are complete
3. Clear separation of concerns - Each item focuses on one specific enhancement rather than trying to do everything

Final Recommendation

• Rating: ⭐⭐⭐⭐☆ (4/5)
• Action: APPROVE
• Reasoning: This is high-quality documentation work that significantly improves backlog organization. The dependency tracking system transforms a flat feature list into a structured implementation graph. All dependency references are valid, markdown structure is consistent, and the tier organization follows logical implementation patterns. The only reason for 4/5 instead of 5/5 is that it's documentation rather than functional code, but within its domain this is exemplary work.

The PR successfully delivers on its stated goals: adds Tier 1g with 12 new visualization items, introduces dependency tracking across all tables, and improves organizational clarity without any structural issues.

---

[greptile-apps]

Greptile Summary

This PR adds Tier 1g — Visualization leverage (items #59–70) to the backlog and retrofits a Depends on column across every existing table, making dependency chains explicit and scannable. It is a docs-only change with no functional code modifications.

Key changes:

• 4 foundation items (docs: update performance benchmarks #59 scoped export, docs: update performance benchmarks #60 symbol-level edge labels, fix: run benchmark after publish to prevent workflow cancellation #61 node export annotations, docs: reorganize README — lead with problem and value #62 semantic subgraph labels) that establish shared visualization infrastructure
• 8 downstream integration items (perf: reduce WASM boundary crossings in JS extractor #63–70) that apply each foundation to specific tools (audit, context, diff-impact, branch-compare, deps, sequence, communities, triage)
• Depends on column added to every backlog table with verified references — all dependency IDs point to valid existing items

Minor prose fix needed: Line 121 incorrectly groups plot with Mermaid/DOT producers, but item #59 correctly identifies it as an interactive HTML viewer via vis-network. Only five commands produce Mermaid/DOT output; plot is an HTML renderer that benefits from the visualization foundations.

Confidence Score: 4/5

• Docs-only change safe to merge after fixing the one prose inaccuracy on line 121 about plot's output format.
• This PR is a low-risk documentation update with no functional code changes. All dependency IDs are valid, table structure is consistent, and backlog items are well-structured. One verified prose inaccuracy exists: line 121 incorrectly groups plot with Mermaid/DOT producers when it actually produces interactive HTML (contradicting item docs: update performance benchmarks #59). This is straightforward to fix and the PR should not merge until corrected.
• docs/roadmap/BACKLOG.md line 121 — prose inaccuracy about plot's output format needs the suggested fix.

Flowchart

%%{init: {'theme': 'neutral'}}%%
flowchart TD
    subgraph Foundation["Tier 1g — Foundation Items"]
        59["#59 Scoped graph rendering"]
        60["#60 Symbol-level edge labels"]
        61["#61 Node export annotations"]
        62["#62 Semantic subgraph labels"]
    end

    subgraph Downstream["Downstream Integration Items"]
        63["#63 Scoped diagram in audit --diagram"]
        64["#64 Scoped diagram in context --diagram"]
        65["#65 Symbol labels in diff-impact mermaid"]
        66["#66 Symbol labels in branch-compare mermaid"]
        67["#67 Symbol-level imports in deps"]
        68["#68 Symbol-level messages in sequence"]
        69["#69 Node annotations in diff-impact / branch-compare / communities"]
        70["#70 Drift/risk subgraph labels in communities and triage"]
    end

    59 --> 62
    59 --> 63
    59 --> 64
    60 --> 65
    60 --> 66
    60 --> 67
    60 --> 68
    61 --> 69
    62 --> 70

Loading

_{Last reviewed commit: d085abf}

[greptile-apps]

plot incorrectly listed as a Mermaid/DOT producer

The Tier 1g intro states "Six commands already produce Mermaid/DOT output: export, diff-impact -f mermaid, branch-compare --format mermaid, sequence, cfg, and plot", but plot produces interactive HTML via vis-network, not Mermaid/DOT. Item #59 on line 125 correctly identifies it as an "interactive HTML viewer" — this intro contradicts that.

Only five commands produce Mermaid/DOT output. plot is a recipient of #59's scoped filtering (HTML side), not a Mermaid/DOT emitter.

Suggested change

	Six commands already produce Mermaid/DOT output: `export`, `diff-impact -f mermaid`, `branch-compare --format mermaid`, `sequence`, `cfg`, and `plot`. These features enrich the shared visualization pipeline and underlying data model so all visual tools produce richer, PR-ready architectural diagrams.
	Six commands already produce visual output: `export` (Mermaid/DOT), `diff-impact -f mermaid`, `branch-compare --format mermaid`, `sequence`, `cfg` (all Mermaid), and `plot` (interactive HTML). These features enrich the shared visualization pipeline and underlying data model so all visual tools produce richer, PR-ready architectural diagrams.