feat(captcha): best-effort rotation/funcaptcha solver

This commit is contained in:
2026-06-07 13:17:26 +01:00
parent 6e496eeac8
commit 13ff6449a3
2 changed files with 54 additions and 0 deletions
+35
View File
@@ -0,0 +1,35 @@
package captcha
import (
"context"
"encoding/json"
"fmt"
"github.com/lukaszraczylo/captcha-solver-mcp/internal/llm"
)
// RotationResult is a best-effort FunCaptcha/rotation answer.
type RotationResult struct {
Choice int `json:"choice"`
Rotations int `json:"rotations"`
Confidence float64 `json:"confidence"`
}
// SolveRotation handles FunCaptcha-style rotation/choice puzzles (best-effort).
func SolveRotation(ctx context.Context, c llm.Client, img []byte, mediaType, instruction string) (RotationResult, error) {
prompt := fmt.Sprintf(
"You are solving a rotation/choice puzzle captcha. Instruction: %q. "+
"If asked to pick one image, return its 0-based index as choice. "+
"If asked to rotate, return the number of clockwise steps as rotations. "+
`Respond ONLY with compact JSON: {"choice":<int>,"rotations":<int>,"confidence":<0..1>}.`,
instruction)
raw, err := c.Vision(ctx, prompt, []llm.Image{{MediaType: mediaType, Data: img}}, llm.Options{})
if err != nil {
return RotationResult{}, err
}
var res RotationResult
if err := json.Unmarshal([]byte(extractJSON(raw)), &res); err != nil {
return RotationResult{}, fmt.Errorf("model did not return parseable rotation answer")
}
return res, nil
}
+19
View File
@@ -0,0 +1,19 @@
package captcha
import (
"context"
"testing"
"github.com/lukaszraczylo/captcha-solver-mcp/internal/llm"
)
func TestSolveRotation(t *testing.T) {
f := &llm.Fake{VisionResp: `{"choice":3,"rotations":2,"confidence":0.6}`}
res, err := SolveRotation(context.Background(), f, []byte{1}, "image/png", "rotate the animal upright")
if err != nil {
t.Fatal(err)
}
if res.Choice != 3 || res.Rotations != 2 {
t.Fatalf("got %+v", res)
}
}