eric/weatherreporter
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases 2 Wiki Activity

Prepare reports for Promptkit migration

Browse Source
This commit is contained in:
Eric Rakestraw
2026-07-31 03:53:47 +00:00
parent 2e0fb65a8b
commit d5fcbfd20c
87 changed files with 2223 additions and 3382 deletions
Download Patch File Download Diff File Expand all files Collapse all files

38
internal/reporttemplate/prompts/daily.generated_text.md
View File

@@ -1,38 +0,0 @@
TASK: You are writing structured prose slots for a dated daily weather report.
The calling application will render the final Markdown report. Your job is not to write the full report. Return only a JSON object matching the configured schema.
Use only the supplied `data_package`. Do not invent weather details, times, hazards, probabilities, or impacts that are not supported by the data.
The report focuses on the selected local civil day in `report.valid_period` for the configured location.
Return these fields:
- `summary`: required. 1-2 sentences summarizing the main weather story for the selected day.
- `forecast_discussion`: required. 1 or more short paragraphs explaining the setup, timing, trend, or uncertainty most relevant to the selected day.
- `precipitation_timing`: optional. Include only when the deterministic `precip_timing` module contains precipitation windows.
- `confidence`: optional. Include only if uncertainty, timing spread, or conflicting signals materially affect how the reader should interpret the forecast.
Return JSON only.
# summary
Lead with the most practical local outcome for the selected day. If an active warning is relevant during the report period, lead with the hazard.
Mention the expected temperature character, precipitation risk, wind, visibility, heat, cold, or other hazards only when supported by the data package.
# forecast_discussion
Use deterministic module facts and narrative products to explain the most useful details for the selected day.
Useful context may include:
- timing of condition changes by daypart or hour
- boundaries, forcing, moisture, instability, or storm mode when supported
- active alerts or SPC outlooks that apply to the location
- planning concerns surfaced by `daily_planning`
- confidence or uncertainty
# precipitation_timing
Use 1-2 sentences to add practical precipitation context only when the data package contains deterministic precipitation windows. Include expected timing, type, intensity, duration, and uncertainty only when those details are supported.

56
internal/reporttemplate/prompts/hourly.generated_text.md
View File

@@ -1,56 +0,0 @@
TASK: You are writing structured prose slots for a short-term hourly weather report.
The calling application will render the final Markdown report. Your job is not to write the full report. Return only a JSON object matching the configured schema.
Use only the supplied `data_package`. Do not invent weather details, times, hazards, probabilities, or impacts that are not supported by the data.
The report focuses on the valid period in `report.valid_period`, typically the next several hours for the configured location.
Return these fields:
- `summary`: required. 1-2 sentences summarizing the main weather story for the valid period.
- `forecast_discussion`: required. 2-3 sentences explaining the broader setup, trend, or forecast reasoning most relevant to the valid period.
- `precipitation_timing`: optional. Include only when the deterministic `precip_timing` module contains precipitation windows.
- `confidence`: optional. Include only if uncertainty, timing spread, or conflicting signals materially affect how the reader should interpret the forecast.
Return JSON only.
# summary
The summary should typically consist of two sentences.
If an active warning is relevant during the report period, lead with the hazard. Otherwise, the first sentence should state the most likely local weather outcome for the valid period, including the overall character of the weather and expected temperature/temperature range.
If the forecast indicates a significant shift in conditions over time (e.g., from sunny to overcast), then identify the hour when the shift is most likely to occur. If the conditions are generally similar or stable across the forecast period, then pick a single descriptor (e.g., mostly clear) that best captures the character of the weather.
The second sentence should state the most important active hazard, caveat, uncertainty, or alternate outcome, if one exists. If there is no meaningful caveat, the second sentence may be omitted, or may briefly say that no major complications are apparent.
In the lead, distinguish the main weather outcome from the caveat. If showers and thunderstorms have different timing, state that difference rather than combining them as a single risk throughout the valid period. If the main caveat is a regional severe-weather or precipitation risk displaced from the report location, state that limitation clearly.
Example style:
- “The rest of the afternoon is expected to be warm and dry, with mostly clear skies. There is a slight chance of isolated showers and thunderstorms developing from late afternoon into early evening.”
# forecast_discussion
Use narrative products to explain the “why” behind the local forecast when useful.
Useful context may include:
- synoptic pattern
- fronts or boundaries
- shortwaves, troughs, or ridges
- instability, moisture, shear, forcing, or capping
- regional placement of precipitation or severe-weather chances
- hazard types and timing windows
- confidence or uncertainty
- conditional outcomes
- relevant notes about the following day or days
# precipitation_timing
Use 1-2 sentences to add practical context, including:
- Whether the precipitation is associated with a moving frontal boundary, convective initiation, or wide stratiform rain (if this can be determined from the data package);
- The expected type, intensity, and duration of the precipitation; and
- Any caveats or uncertainty with respect to the onset, duration, or occurrance of the precipitation.

38
internal/reporttemplate/prompts/today.generated_text.md
View File

@@ -1,38 +0,0 @@
TASK: You are writing structured prose slots for a current-day weather report.
The calling application will render the final Markdown report. Your job is not to write the full report. Return only a JSON object matching the configured schema.
Use only the supplied `data_package`. Do not invent weather details, times, hazards, probabilities, or impacts that are not supported by the data.
The report focuses on today's valid period in `report.valid_period` for the configured location.
Return these fields:
- `summary`: required. 1-2 sentences summarizing the main weather story for today.
- `forecast_discussion`: required. 1 or more short paragraphs explaining the setup, timing, trend, or uncertainty most relevant to today.
- `precipitation_timing`: optional. Include only when the deterministic `precip_timing` module contains precipitation windows.
- `confidence`: optional. Include only if uncertainty, timing spread, or conflicting signals materially affect how the reader should interpret the forecast.
Return JSON only.
# summary
Lead with the most practical local outcome for today. If an active warning is relevant during the report period, lead with the hazard.
Mention the expected temperature character, precipitation risk, wind, visibility, heat, cold, or other hazards only when supported by the data package.
# forecast_discussion
Use deterministic module facts and narrative products to explain the most useful details for the current day.
Useful context may include:
- timing of condition changes by daypart or hour
- boundaries, forcing, moisture, instability, or storm mode when supported
- active alerts or SPC outlooks that apply to the location
- planning concerns surfaced by `today_planning`
- confidence or uncertainty
# precipitation_timing
Use 1-2 sentences to add practical precipitation context only when the data package contains deterministic precipitation windows. Include expected timing, type, intensity, duration, and uncertainty only when those details are supported.

54
internal/reporttemplate/prompts/tomorrow.generated_text.md
View File

@@ -1,54 +0,0 @@
TASK: You are writing structured prose slots for a short-term hourly weather report.
The calling application will render the final Markdown report. Your job is not to write the full report. Return only a JSON object matching the configured schema.
Use only the supplied `data_package`. Do not invent weather details, times, hazards, probabilities, or impacts that are not supported by the data.
The report focuses on the valid period in `report.valid_period`, typically the next several hours for the configured location.
Return these fields:
- `summary`: required. 1-2 sentences summarizing the main weather story for the valid period.
- `forecast_discussion`: required. 2-3 sentences explaining the broader setup, trend, or forecast reasoning most relevant to the valid period.
- `precipitation_timing`: optional. Include only when the deterministic `precip_timing` module contains precipitation windows.
- `confidence`: optional. Include only if uncertainty, timing spread, or conflicting signals materially affect how the reader should interpret the forecast.
Return JSON only.
# summary
The summary should typically consist of two sentences.
If an active warning is relevant during the report period, lead with the hazard. Otherwise, the first sentence should state the most likely local weather outcome for the valid period, including the overall character of the weather and expected temperature/temperature range.
The second sentence should state the most important active hazard, caveat, uncertainty, or alternate outcome, if one exists. If there is no meaningful caveat, the second sentence may be omitted, or may briefly say that no major complications are apparent.
In the lead, distinguish the main weather outcome from the caveat. If showers and thunderstorms have different timing, state that difference rather than combining them as a single risk throughout the valid period. If the main caveat is a regional severe-weather or precipitation risk displaced from the report location, state that limitation clearly.
Example style:
- “Sunday is expected to be warm and dry, with mostly clear skies. There is a slight chance of isolated showers and thunderstorms developing from late afternoon into early evening.”
# forecast_discussion
Use narrative products to explain the “why” behind the local forecast when useful.
Useful context may include:
- synoptic pattern
- fronts or boundaries
- shortwaves, troughs, or ridges
- instability, moisture, shear, forcing, or capping
- regional placement of precipitation or severe-weather chances
- hazard types and timing windows
- confidence or uncertainty
- conditional outcomes
- relevant notes about the following day or days
# precipitation_timing
Use 1-2 sentences to add practical context, including:
- Whether the precipitation is associated with a moving frontal boundary, convective initiation, or wide stratiform rain (if this can be determined from the data package);
- The expected type, intensity, and duration of the precipitation; and
- Any caveats or uncertainty with respect to the onset, duration, or occurrance of the precipitation.

23
internal/reporttemplate/reporttemplate.go
View File

@@ -1,4 +1,4 @@
// Package reporttemplate provides embedded Markdown templates and schemas.
// Package reporttemplate provides embedded Markdown templates and partials.
package reporttemplate
import (
@@ -8,7 +8,7 @@ import (
"text/template"
)
//go:embed templates/*.md.tmpl templates/partials/*.md.tmpl schemas/*.schema.json
//go:embed templates/*.md.tmpl templates/partials/*.md.tmpl
var assets embed.FS
var templates = map[string]string{
@@ -25,13 +25,6 @@ var templatePartials = []string{
"templates/partials/today_daypart_forecast.md.tmpl",
}
var schemas = map[string]string{
"daily": "schemas/daily.generated_text.schema.json",
"hourly": "schemas/hourly.generated_text.schema.json",
"today": "schemas/today.generated_text.schema.json",
"tomorrow": "schemas/tomorrow.generated_text.schema.json",
}
func Template(id string) (string, error) {
path, ok := templates[id]
if !ok {
@@ -44,18 +37,6 @@ func Template(id string) (string, error) {
return string(data), nil
}
func Schema(id string) ([]byte, error) {
path, ok := schemas[id]
if !ok {
return nil, fmt.Errorf("unknown generated text schema %q", id)
}
data, err := assets.ReadFile(path)
if err != nil {
return nil, fmt.Errorf("read generated text schema %q: %w", id, err)
}
return append([]byte(nil), data...), nil
}
func Render(id string, data any) ([]byte, error) {
source, err := Template(id)
if err != nil {

26
internal/reporttemplate/reporttemplate_test.go
View File

@@ -2,9 +2,10 @@ package reporttemplate
import (
"encoding/json"
"os"
"strings"
"testing"
"gitea.maximumdirect.net/eric/weatherreporter/internal/promptassets"
)
func TestTemplateLookup(t *testing.T) {
@@ -59,7 +60,7 @@ func TestTodayTemplateLookup(t *testing.T) {
}
func TestSchemaLookup(t *testing.T) {
data, err := Schema("hourly")
data, err := promptassets.Schema("hourly")
if err != nil {
t.Fatalf("Schema() error = %v", err)
}
@@ -67,7 +68,7 @@ func TestSchemaLookup(t *testing.T) {
}
func TestTomorrowSchemaLookup(t *testing.T) {
data, err := Schema("tomorrow")
data, err := promptassets.Schema("tomorrow")
if err != nil {
t.Fatalf("Schema() error = %v", err)
}
@@ -98,7 +99,7 @@ func TestTomorrowSchemaLookup(t *testing.T) {
}
func TestDailySchemaLookup(t *testing.T) {
data, err := Schema("daily")
data, err := promptassets.Schema("daily")
if err != nil {
t.Fatalf("Schema() error = %v", err)
}
@@ -129,7 +130,7 @@ func TestDailySchemaLookup(t *testing.T) {
}
func TestTodaySchemaLookup(t *testing.T) {
data, err := Schema("today")
data, err := promptassets.Schema("today")
if err != nil {
t.Fatalf("Schema() error = %v", err)
}
@@ -904,26 +905,11 @@ func TestUnknownAssetsReturnActionableErrors(t *testing.T) {
if _, err := Template("missing"); err == nil || !strings.Contains(err.Error(), `unknown report template "missing"`) {
t.Fatalf("Template() error = %v, want unknown template", err)
}
if _, err := Schema("missing"); err == nil || !strings.Contains(err.Error(), `unknown generated text schema "missing"`) {
t.Fatalf("Schema() error = %v, want unknown schema", err)
}
if _, err := Render("missing", testRenderContext{}); err == nil || !strings.Contains(err.Error(), `unknown report template "missing"`) {
t.Fatalf("Render() error = %v, want unknown template", err)
}
}
func TestDailyPromptAssetExists(t *testing.T) {
data, err := os.ReadFile("prompts/daily.generated_text.md")
if err != nil {
t.Fatalf("read Daily prompt asset: %v", err)
}
for _, want := range []string{"TASK:", "`summary`", "`forecast_discussion`", "`daily_planning`"} {
if !strings.Contains(string(data), want) {
t.Fatalf("Daily prompt asset missing %q:\n%s", want, string(data))
}
}
}
func TestRenderFailsForMissingContextFields(t *testing.T) {
_, err := Render("hourly", map[string]any{"Report": map[string]any{"Title": "Hourly Report"}})
if err == nil {

29
internal/reporttemplate/schemas/daily.generated_text.schema.json
View File

@@ -1,29 +0,0 @@
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "weatherreporter.daily.generated_text.schema.json",
"title": "Daily GeneratedText",
"type": "object",
"additionalProperties": false,
"required": [
"summary",
"forecast_discussion"
],
"properties": {
"summary": {
"type": "string"
},
"forecast_discussion": {
"type": "array",
"items": {
"type": "string"
},
"minItems": 1
},
"precipitation_timing": {
"type": "string"
},
"confidence": {
"type": "string"
}
}
}

25
internal/reporttemplate/schemas/hourly.generated_text.schema.json
View File

@@ -1,25 +0,0 @@
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "weatherreporter.hourly.generated_text.schema.json",
"title": "Hourly GeneratedText",
"type": "object",
"additionalProperties": false,
"required": [
"summary",
"forecast_discussion"
],
"properties": {
"summary": {
"type": "string"
},
"forecast_discussion": {
"type": "string"
},
"precipitation_timing": {
"type": "string"
},
"confidence": {
"type": "string"
}
}
}

29
internal/reporttemplate/schemas/today.generated_text.schema.json
View File

@@ -1,29 +0,0 @@
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "weatherreporter.today.generated_text.schema.json",
"title": "Today GeneratedText",
"type": "object",
"additionalProperties": false,
"required": [
"summary",
"forecast_discussion"
],
"properties": {
"summary": {
"type": "string"
},
"forecast_discussion": {
"type": "array",
"items": {
"type": "string"
},
"minItems": 1
},
"precipitation_timing": {
"type": "string"
},
"confidence": {
"type": "string"
}
}
}

29
internal/reporttemplate/schemas/tomorrow.generated_text.schema.json
View File

@@ -1,29 +0,0 @@
{
"$schema": "https://json-schema.org/draft/2020-12/schema",
"$id": "weatherreporter.tomorrow.generated_text.schema.json",
"title": "Tomorrow GeneratedText",
"type": "object",
"additionalProperties": false,
"required": [
"summary",
"forecast_discussion"
],
"properties": {
"summary": {
"type": "string"
},
"forecast_discussion": {
"type": "array",
"items": {
"type": "string"
},
"minItems": 1
},
"precipitation_timing": {
"type": "string"
},
"confidence": {
"type": "string"
}
}
}