Extract agent and tools from tracker mapping worker into dedicated files

Split tracker_mapping_worker.go: agent construction, prompts, and
structured output type move to tracker_mapping_agent.go; each tool gets
its own *_tool.go file. Update naming conventions (worker, agent, tool
file patterns, AgentResult suffix, RunTyped preference).

Signed-off-by: Émile Ré <emile@probo.com>
This commit is contained in:
Émile Ré
2026-05-19 12:06:15 +04:00
parent 10adf2bd4b
commit c4147e6801
7 changed files with 256 additions and 174 deletions

View File

@@ -16,45 +16,26 @@ package cookiebanner
import (
"context"
_ "embed"
"errors"
"fmt"
"strings"
"time"
"go.gearno.de/kit/log"
"go.gearno.de/kit/pg"
"go.gearno.de/kit/worker"
"go.probo.inc/probo/pkg/agent"
"go.probo.inc/probo/pkg/agent/tools/search"
"go.probo.inc/probo/pkg/coredata"
"go.probo.inc/probo/pkg/gid"
"go.probo.inc/probo/pkg/llm"
"go.probo.inc/probo/pkg/slug"
)
const (
agentTimeout = 60 * time.Second
agentMaxTurns = 5
agentConfidenceThreshold = 0.6
agentMaxPatternConfidence = 0.8
)
//go:embed prompts/tracker_identification.txt.tmpl
var trackerIdentificationPrompt string
type trackerMappingHandler struct {
pg *pg.Client
logger *log.Logger
agent *agent.Agent
}
type TrackerMappingConfig struct {
LLMClient *llm.Client
Model string
FirecrawlAPIKey string
}
func NewTrackerMappingWorker(
pgClient *pg.Client,
logger *log.Logger,
@@ -78,52 +59,6 @@ func NewTrackerMappingWorker(
)
}
func buildTrackerMappingAgent(
cfg TrackerMappingConfig,
pgClient *pg.Client,
logger *log.Logger,
) *agent.Agent {
tools := []agent.Tool{
searchTrackerPatternsTool(pgClient),
searchThirdPartiesTool(pgClient),
}
if cfg.FirecrawlAPIKey != "" {
tools = append(tools, search.FirecrawlSearchTool(cfg.FirecrawlAPIKey))
}
outputType, err := agent.NewOutputType[TrackerIdentification]("tracker_identification")
if err != nil {
panic(fmt.Sprintf("cookiebanner: cannot build tracker identification output type: %s", err))
}
return agent.New(
"tracker-mapping",
cfg.LLMClient,
agent.WithInstructionsFunc(trackerMappingInstructions),
agent.WithModel(cfg.Model),
agent.WithTools(tools...),
agent.WithOutputType(outputType),
agent.WithMaxTurns(agentMaxTurns),
agent.WithLogger(logger),
)
}
func trackerMappingInstructions(_ context.Context, _ *agent.Agent) string {
categories := coredata.ThirdPartyCategories()
parts := make([]string, len(categories))
for i, c := range categories {
parts[i] = string(c)
}
return strings.Replace(
trackerIdentificationPrompt,
"{{.Categories}}",
strings.Join(parts, ", "),
1,
)
}
func (h *trackerMappingHandler) Claim(ctx context.Context) (coredata.TrackerPattern, error) {
var tp coredata.TrackerPattern
@@ -292,7 +227,7 @@ func (h *trackerMappingHandler) identifyWithAgent(
agentCtx, cancel := context.WithTimeout(ctx, agentTimeout)
defer cancel()
result, err := agent.RunTyped[TrackerIdentification](
result, err := agent.RunTyped[TrackerMappingAgentResult](
agentCtx,
h.agent,
[]llm.Message{
@@ -379,7 +314,7 @@ func (h *trackerMappingHandler) identifyWithAgent(
func (h *trackerMappingHandler) resolveOrCreateCommonThirdParty(
ctx context.Context,
tx pg.Tx,
identification TrackerIdentification,
identification TrackerMappingAgentResult,
domains []string,
) (*gid.GID, error) {
var party coredata.CommonThirdParty
@@ -490,28 +425,3 @@ func (h *trackerMappingHandler) resolveThirdParty(
return &t.ID, nil
}
func buildAgentPrompt(tp coredata.TrackerPattern, domains []string) string {
maxAge := "session"
if tp.MaxAgeSeconds != nil {
maxAge = fmt.Sprintf("%d seconds", *tp.MaxAgeSeconds)
}
prompt := fmt.Sprintf(
"Identify the following tracker:\n\n"+
"<pattern> %s </pattern>\n"+
"<type> %s </type>\n"+
"<match_type> %s </match_type>\n"+
"<max_age> %s </max_age>\n",
tp.Pattern,
tp.TrackerType,
tp.MatchType,
maxAge,
)
if len(domains) > 0 {
prompt += fmt.Sprintf("<observed_domains> %s </observed_domains>\n", strings.Join(domains, ", "))
}
return prompt
}