Feedback
When running zcodegraph index on jununfly/ZAgenticLoop, the index succeeds and is usable, but the fallback health is reported as degraded. The top-level CLI output says:
Rust-hybrid appended 30 TypeScript fallback files
Fallback health: degraded
Top fallback reasons:
30 TypeScript fallback files
However, the diagnostic bundle's structured fields classify the fallback as YAML files handled by the TypeScript fallback path:
{
"fallbackState": "degraded",
"fallbackFileCount": 30,
"fallbackReasonTaxonomy": {
"language-level-typescript-fallback": 30
},
"fallbackByLanguage": {
"yaml": 30
},
"fallbackMessage": "TypeScript fallback appended 30 non-Rust-owned supported file(s)."
}
This is technically understandable after reading the bundle, but the CLI summary is easy to misread as "30 TypeScript source files fell back" rather than "30 YAML files were indexed through the TypeScript fallback implementation."
Reproduction
Repository:
Commands:
zcodegraph index
zcodegraph doctor --engine rust-hybrid --bundle --last-run
Observed index summary:
Indexed 52 files
719 nodes, 1,255 edges
Indexed with rust-hybrid
Fallback health: degraded
Top fallback reasons:
30 TypeScript fallback files
Diagnostic bundle generated locally:
.zcodegraph/diagnostics/bundles/2026-07-08T03-08-21-148Z-last-run
Replay-safe bundle summary:
{
"fileCount": 52,
"nodeCount": 719,
"edgeCount": 1255,
"filesByLanguage": {
"python": 1,
"typescript": 21,
"yaml": 30
},
"fallbackByLanguage": {
"yaml": 30
},
"fallbackReasonTaxonomy": {
"language-level-typescript-fallback": 30
},
"missingFallbackFileCount": 0
}
Expected improvement
Could the CLI and doctor summary distinguish fallback implementation from source language more clearly?
For example:
30 YAML files indexed via TypeScript fallback
- or
30 non-Rust-owned supported files used TypeScript fallback (yaml: 30)
This would make the degraded fallback report easier for users to interpret and decide whether action is needed.
Impact
The index is usable, so this is not a blocker. It is mainly a diagnostics clarity issue that affects user confidence and triage speed.
Feedback
When running
zcodegraph indexonjununfly/ZAgenticLoop, the index succeeds and is usable, but the fallback health is reported as degraded. The top-level CLI output says:However, the diagnostic bundle's structured fields classify the fallback as YAML files handled by the TypeScript fallback path:
{ "fallbackState": "degraded", "fallbackFileCount": 30, "fallbackReasonTaxonomy": { "language-level-typescript-fallback": 30 }, "fallbackByLanguage": { "yaml": 30 }, "fallbackMessage": "TypeScript fallback appended 30 non-Rust-owned supported file(s)." }This is technically understandable after reading the bundle, but the CLI summary is easy to misread as "30 TypeScript source files fell back" rather than "30 YAML files were indexed through the TypeScript fallback implementation."
Reproduction
Repository:
jununfly/ZAgenticLoopCommands:
Observed index summary:
Diagnostic bundle generated locally:
Replay-safe bundle summary:
{ "fileCount": 52, "nodeCount": 719, "edgeCount": 1255, "filesByLanguage": { "python": 1, "typescript": 21, "yaml": 30 }, "fallbackByLanguage": { "yaml": 30 }, "fallbackReasonTaxonomy": { "language-level-typescript-fallback": 30 }, "missingFallbackFileCount": 0 }Expected improvement
Could the CLI and doctor summary distinguish fallback implementation from source language more clearly?
For example:
30 YAML files indexed via TypeScript fallback30 non-Rust-owned supported files used TypeScript fallback (yaml: 30)This would make the degraded fallback report easier for users to interpret and decide whether action is needed.
Impact
The index is usable, so this is not a blocker. It is mainly a diagnostics clarity issue that affects user confidence and triage speed.