opencode-see-image 0.6.1 → 0.7.0

package/README.md

@@ -102,6 +102,7 @@ All settings are env-var overrides. The plugin uses opencode's SDK client by def
 | `SEE_IMAGE_ENDPOINT` | `https://opencode.ai/zen/go/v1/messages` | HTTP endpoint (only used if `SEE_IMAGE_API_KEY` is set) |
 | `SEE_IMAGE_API_VERSION` | `2023-06-01` | `anthropic-version` header (HTTP mode only) |
 | `SEE_IMAGE_USER_AGENT` | _(Chrome UA)_ | User-Agent header (HTTP mode only) |
+| `SEE_IMAGE_TIMEOUT` | `30000` | Per-candidate timeout in ms. Prevents hanging on slow models. |
 
 ### Using a different vision model
 

package/index.ts

@@ -11,6 +11,7 @@ const ENDPOINT =
   "https://opencode.ai/zen/go/v1/messages"
 const MODEL = process.env.SEE_IMAGE_MODEL || "minimax-m3"
 const PROVIDER_ID = process.env.SEE_IMAGE_PROVIDER || "opencode-go"
+const TIMEOUT = parseInt(process.env.SEE_IMAGE_TIMEOUT || "30000", 10)
 const API_VERSION = process.env.SEE_IMAGE_API_VERSION || "2023-06-01"
 const USER_AGENT =
   process.env.SEE_IMAGE_USER_AGENT ||
@@ -194,6 +195,26 @@ function resolveImage(name: string, cwd: string, sessionID?: string): ResolvedIm
   )
 }
 
+function readProviderKey(providerID: string): string | null {
+  try {
+    const xdgDataHome = process.env.XDG_DATA_HOME
+      ? path.join(process.env.XDG_DATA_HOME, "opencode")
+      : ""
+    const dataDir =
+      process.env.OPENCODE_DATA_DIR ||
+      xdgDataHome ||
+      path.join(os.homedir(), ".local/share/opencode")
+    const authPath = path.join(dataDir, "auth.json")
+    if (!fs.existsSync(authPath)) return null
+    const auth = JSON.parse(fs.readFileSync(authPath, "utf8"))
+    const entry = auth[providerID]
+    if (entry?.type === "api" && entry?.key) return entry.key
+    return null
+  } catch {
+    return null
+  }
+}
+
 async function seeImageViaSDK(
   client: any,
   dataUrl: string,
@@ -222,6 +243,10 @@ async function seeImageViaSDK(
         continue
       }
 
+      // Per-candidate timeout so a slow model doesn't hang forever
+      const controller = new AbortController()
+      const timer = setTimeout(() => controller.abort(), TIMEOUT)
+
       const result = await client.session.prompt({
         path: { id: sessionID },
         body: {
@@ -234,7 +259,9 @@ async function seeImageViaSDK(
           system:
             "You are a vision assistant. Describe the image accurately and concisely. Answer with text only.",
         },
+        signal: controller.signal,
       })
+      clearTimeout(timer)
 
       const parts = result.data?.parts ?? []
       const text = (parts as any[])
@@ -259,6 +286,18 @@ async function seeImageViaSDK(
     }
   }
 
+  // If user has an API key configured (from auth.json or env), try HTTP fallback
+  const b64 = dataUrl.split(",")[1] || ""
+  const apiKey =
+    process.env.SEE_IMAGE_API_KEY || readProviderKey("opencode-go")
+  if (apiKey) {
+    try {
+      return await seeImageViaHTTP(b64, mediaType, prompt, abort, apiKey)
+    } catch (e: any) {
+      errors.push(`http-fallback: ${e?.message ?? e}`)
+    }
+  }
+
   throw new Error(
     `see_image: SDK vision call failed for all candidates. ${errors.join("; ")}`,
   )
@@ -269,8 +308,9 @@ async function seeImageViaHTTP(
   mediaType: string,
   prompt: string,
   abort?: AbortSignal,
+  keyOverride?: string,
 ): Promise<{ text: string; model: string; provider: string }> {
-  const key = process.env.SEE_IMAGE_API_KEY!
+  const key = keyOverride || process.env.SEE_IMAGE_API_KEY!
   const body = {
     model: MODEL,
     max_tokens: 2048,

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-see-image",
-  "version": "0.6.1",
+  "version": "0.7.0",
   "description": "Give non-vision opencode models the ability to see images/screenshots by routing them to a vision-capable model (MiniMax M3 via opencode-go by default).",
   "type": "module",
   "main": "index.ts",