tsa-temporallisted
Install: claude install-skill aimasteracc/tree-sitter-analyzer
# tsa-temporal — Hot zones via git history
> Per-symbol modification frequency persisted in `ast_symbol_activation`.
> Computed from `git log --follow -p -U0` hunk attribution at index time.
> Symbols with `mod_count_30d >= 5` auto-trigger CAUTION in change_impact.
## When to use
| Goal | How |
|---------------------------------------|------------------------------------------------|
| Hot-zone caller fanout | `codegraph_callers(..., include_activation=true)` |
| Hot-zone callee fanout | `codegraph_callees(..., include_activation=true)` |
| "Is this commit touching hot zones?" | `analyze_change_impact` — read `risk_factors` |
| Single-file recent churn | `check_file_health` — read `git_hotspot` dim |
**Don't use** when:
- The question is static (e.g. "who calls X") — use `tsa-graph` skill
- File is brand new (no git history) — temporal data will be all-zero
## Procedure
### Pre-refactor hot-zone scan
For each symbol you're about to refactor:
```
codegraph_callers(
function_name="X",
include_activation=true,
limit=50
)
```
Returns enriched entries:
```yaml
callers: [
{
name: ...,
file: ...,
line: ...,
callee_resolution: ...,
activation: {
mod_count_30d: <int>,
last_modified_at: <unix ts>
}
}, ...
]
```
Filter `mod_count_30d >= 5` to find the callers that have been modified
recently — those are the hig