opencode-see-image 0.6.2 → 0.8.0

package/index.ts

@@ -195,13 +195,74 @@ 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,
   mediaType: string,
   prompt: string,
   abort?: AbortSignal,
+  currentSessionID?: string,
 ): Promise<{ text: string; model: string; provider: string }> {
+  const errors: string[] = []
+
+  // Try current session first (uses its existing model — free big-pickle, no new session needed)
+  if (currentSessionID) {
+    try {
+      const controller = new AbortController()
+      const timer = setTimeout(() => controller.abort(), TIMEOUT)
+      const result = await client.session.prompt({
+        path: { id: currentSessionID },
+        body: {
+          parts: [
+            { type: "file", mime: mediaType, url: dataUrl },
+            { type: "text", text: prompt },
+          ],
+          tools: {},
+          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[])
+        .filter((p: any) => p.type === "text")
+        .map((p: any) => p.text)
+        .filter((t: any) => typeof t === "string" && t.length > 0)
+        .join("\n")
+        .trim()
+
+      if (text) {
+        return { text, model: "", provider: "current" }
+      }
+      errors.push(`current-session: no text in response`)
+    } catch (e: any) {
+      errors.push(`current-session: ${e?.message ?? e}`)
+    }
+  }
+
+  // Fallback: create new sessions with specific provider/model candidates
   const envProvider = process.env.SEE_IMAGE_PROVIDER
   const envModel = process.env.SEE_IMAGE_MODEL
   const candidates: Array<{ providerID: string; modelID: string }> = []
@@ -211,8 +272,6 @@ async function seeImageViaSDK(
   candidates.push({ providerID: "opencode-go", modelID: "minimax-m3" })
   candidates.push({ providerID: "opencode", modelID: "big-pickle" })
 
-  const errors: string[] = []
-
   for (const { providerID, modelID } of candidates) {
     let sessionID: string | undefined
     try {
@@ -266,8 +325,24 @@ 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}`)
+    }
+  }
+
+  const errMsg = errors.join("; ")
+  const hint = errMsg.includes("usage limit")
+    ? ` Enable usage from your balance at https://opencode.ai/workspace/wrk_01KVARG0A0Y87XV5JYBNJ0WRXB/go`
+    : ""
   throw new Error(
-    `see_image: SDK vision call failed for all candidates. ${errors.join("; ")}`,
+    `see_image: SDK vision call failed for all candidates. ${errMsg}.${hint}`,
   )
 }
 
@@ -276,8 +351,9 @@ 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,
@@ -406,6 +482,7 @@ const SeeImagePlugin: Plugin = async (ctx) => {
           resolved.mediaType,
           prompt,
           context.abort,
+          context.sessionID,
         )
       }
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "opencode-see-image",
-  "version": "0.6.2",
+  "version": "0.8.0",
   "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",