xtekky/gpt4free · error · TokenManagerError

TOKEN_ERROR

TOKEN_ERROR

Error message

Failed to get Copilot token: {resp.status} - {text}

What it means

TokenManagerError with code TOKEN_ERROR, raised when the Copilot token exchange returns any non-200/non-401 status. The status code and response body are embedded in the message, covering server-side and unexpected-client errors (403 rate limit/entitlement, 404 bad headers, 5xx outages) during the OAuth-to-Copilot-token exchange.

Source

Thrown at g4f/Provider/github/copilotTokenProvider.py:76

                self.COPILOT_TOKEN_URL,
                headers={
                    "Authorization": f"token {github_token}",
                    "Accept": "application/json",
                    "User-Agent": USER_AGENT,
                    "Editor-Version": EDITOR_VERSION,
                    "Editor-Plugin-Version": EDITOR_PLUGIN_VERSION,
                    "Openai-Organization": "github-copilot",
                    "X-GitHub-Api-Version": API_VERSION,
                },
            ) as resp:
                if resp.status == 401:
                    raise TokenManagerError(
                        "AUTH_FAILED",
                        "GitHub token is invalid or expired. Please login again.",
                    )
                if resp.status != 200:
                    text = await resp.text()
                    raise TokenManagerError(
                        "TOKEN_ERROR",
                        f"Failed to get Copilot token: {resp.status} - {text}",
                    )

                data = await resp.json()
                self._copilot_token = data.get("token")

                # Parse expiration
                expires_at = data.get("expires_at")
                if expires_at:
                    try:
                        # Parse ISO format datetime
                        from datetime import datetime

                        dt = datetime.fromisoformat(expires_at.replace("Z", "+00:00"))
                        self._copilot_token_expires_at = dt.timestamp()
                    except Exception:
                        # Default to 30 minutes from now if parsing fails

View on GitHub (pinned to 973504e177)

Solutions

1. Inspect the embedded status and body text: 403 usually means the account has no Copilot access — verify the subscription
2. Back off and retry for rate-limit/5xx responses; the exchange endpoint is rate-limited
3. Update g4f to the latest version so editor headers stay current

try:
    token = await provider.get_valid_token()
except TokenManagerError as e:
    if e.code == 'TOKEN_ERROR' and (' 429 ' in str(e) or ' 5' in str(e)[:40]):
        await asyncio.sleep(backoff())
        token = await provider.get_valid_token()
    raise

When it happens

Trigger: GET copilot_internal/token returning 403 (account lacks Copilot entitlement or rate-limited), 404 (editor-version headers rejected), or 5xx (GitHub incident) while exchanging a structurally valid GitHub token.

Common situations: Copilot subscription not active on the GitHub account; GitHub secondary rate limits during rapid token refreshes; library's EDITOR_VERSION headers outdated after GitHub API changes.

Related errors

• Failed to chat: {response.status} {error_text}
• Failed to fetch Copilot usage: {resp.status} {text}
• AUTH_FAILED
• Cannot create GLM chat session (status {response.status})
• Failed to decode JSON from PhindAi response: {text}

AI-assisted analysis of xtekky/gpt4free@973504e177 (2026-08-14). Data as JSON: /api/errors/83fcab4811becaad. Report an issue: GitHub.