HKUDS/DeepTutor · error · HubError

{self.name}: publish request failed: {exc}

Error message

{self.name}: publish request failed: {exc}

What it means

Raised by ClawHubProvider.publish when the multipart POST to the hub's publish/upload endpoint fails at the transport level (httpx.HTTPError: connection refused, DNS failure, TLS error, timeout). No HTTP response was received.

Source

Thrown at deeptutor/services/skill/hub.py:474

        zip_bytes: bytes,
        token: str,
        fields: dict[str, str],
    ) -> dict[str, Any]:
        """Publish a version via ``POST /skills`` (multipart + bearer token).

        Read-only ClawHub mirrors will reject this; eduhub (same /api/v1 shape)
        accepts it. Surfaces the API's JSON ``error`` message on failure.
        """
        url = f"{self._base_url}/skills"
        try:
            response = self._client.post(
                url,
                data={"slug": slug, "version": version, **fields},
                files={"zip": ("package.zip", zip_bytes, "application/zip")},
                headers={"Authorization": f"Bearer {token}"},
            )
        except httpx.HTTPError as exc:
            raise HubError(f"{self.name}: publish request failed: {exc}") from exc
        if response.status_code >= 400:
            detail = ""
            try:
                detail = str(response.json().get("error") or "")
            except ValueError:
                detail = response.text[:200]
            raise HubError(f"{self.name}: publish failed (HTTP {response.status_code}): {detail}")
        try:
            return response.json()
        except ValueError:
            return {}

    def list_my_skills(self, token: str) -> list[dict[str, Any]]:
        """List the authenticated user's skills via ``GET /skills?owner=me``.

        Each row carries the slug, current ``version`` (latest), the full
        ``versions`` list (for rollback) and the current classification (for
        pre-filling an upgrade's tagging) — see ``buildMySkills`` on the hub.

View on GitHub (pinned to 3e82f13042)

Solutions

1. Verify connectivity: curl -v $BASE_URL (check DNS/TLS/proxy)
2. Increase the httpx client timeout for large skill zips
3. Fix base_url in settings/skill_hubs.json
4. Configure proxy env vars (HTTPS_PROXY) if behind a corporate proxy

resp = httpx.get(BASE.replace('/api',''), timeout=5)
if resp.status_code >= 500: defer_publish()

for attempt in range(3):
    try:
        return hub.publish(...)
    except HubError as e:
        if "publish request failed" in str(e) and attempt < 2:
            time.sleep(2 ** attempt); continue
        raise

When it happens

Trigger: Calling publish(slug, version, token, ...) (via publish_to_hub / `skill publish`) while the hub host is unreachable, DNS fails, the TLS cert is invalid, or the upload of the large zip times out.

Common situations: Wrong base_url; corporate proxy blocking the POST; self-signed cert on a self-hosted hub; slow upload of a big skill zip exceeding the client timeout; offline environment.

Related errors

• Image generation request error: {exc}
• Image generation request error: {exc}
• {self.name}: catalog returned invalid JSON
• {self.name}: publish failed (HTTP {response.status_code}): {
• TTS request error: {exc}

AI-assisted analysis of HKUDS/DeepTutor@3e82f13042 (2026-08-27). Data as JSON: /api/errors/aef17ed161cab62e. Report an issue: GitHub.