mvanhorn/last30days-skill · error · RuntimeError

No X backend is available.

Error message

No X backend is available.

What it means

RuntimeError raised in the X fetch leg when env.x_backend_chain(config) resolves to an empty chain AND no runtime-pinned backend (runtime.x_search_backend) is set. The chain is built from configured backends (e.g. bird session, ScrapeCreators/grok), so an empty chain means the config has no usable X credential at all. It is raised before any fetch is attempted.

Source

Thrown at skills/last30days/scripts/lib/pipeline.py:4208

                f"{public_failure}",
            )
        return merged, {}
    if source == "x":
        # Compile X query from raw_topic (like Reddit/YouTube), not planner's
        # search_query which may contain operator strings like "Rome Italy".
        x_query = raw_topic or topic or subquery.search_query
        ranking_query = subquery.ranking_query

        # One X source, an ordered chain of interchangeable backends. Try the
        # primary; fall through to the next only if it returns nothing or errors.
        chain = env.x_backend_chain(config)
        # Trust an explicit runtime backend as the primary (already resolved as
        # available), keeping the rest of the chain as failover backups.
        pinned = runtime.x_search_backend
        if pinned:
            chain = [pinned] + [b for b in chain if b != pinned]
        if not chain:
            raise RuntimeError("No X backend is available.")
        last_error = ""
        items = []
        used_backend = None
        for i, backend in enumerate(chain):
            items, err = _fetch_x_backend(backend, x_query, from_date, to_date, depth, config)
            if items:
                if i > 0:
                    print(f"[X] primary backend(s) returned nothing; used fallback '{backend}'", file=sys.stderr)
                # Check for auth errors before proceeding to judge-retry
                if last_error:
                    # Fallback succeeded after earlier backend failed. Classify
                    # the original error: if it was AUTH_FAILED (grok revoked),
                    # preserve that state so user gets re-login guidance.
                    prior_state = http.classify_failure(message=last_error)
                    if prior_state == schema.AUTH_FAILED:
                        # Keep AUTH_FAILED visible so host shows re-login hint
                        return items, _outcome_artifact(
                            schema.AUTH_FAILED,

View on GitHub (pinned to c7460f6114)

Solutions

1. Set SCRAPECREATORS_API_KEY (onboarding persists it automatically via setup_wizard.write_api_key).
2. Re-run the setup wizard / GitHub device-auth flow to restore the bird or grok backend.
3. Or exclude 'x' from requested_sources and rely on keyless sources.
4. Check env.x_backend_chain(config) returns non-empty before running if you orchestrate configs dynamically.

Example fix

# before
run_pipeline(topic, requested_sources=["x"], config={})

# after
run_pipeline(topic, requested_sources=["x"], config={"SCRAPECREATORS_API_KEY": os.environ["SCRAPECREATORS_API_KEY"]})

from skills.last30days.scripts.lib import env
if not env.x_backend_chain(config):
    raise SystemExit("no X backend configured: set SCRAPECREATORS_API_KEY or run setup")

try:
    items, meta = fetch_x(...)
except RuntimeError as exc:
    if "No X backend" in str(exc):
        sources.remove("x")
        items, meta = None, None  # continue with remaining sources

When it happens

Trigger: Calling the pipeline with source 'x' while: no ScrapeCreators key, no bird session file, and X_SEARCH_BACKEND env unset/empty; or every backend in the chain has been disabled via config. pinned is falsy, so chain stays [].

Common situations: Fresh install that skipped onboarding (no GitHub device-auth, no SCRAPECREATORS_API_KEY); expired/removed bird session; env var typos like X_BACKEND instead of the expected names; CI environments with no credentials.

Related errors

• No sources are available for this run.
• All X backends failed — {last_error}
• Gemini selected but no Google API key is configured.
• xAI selected but XAI_API_KEY is not configured.
• OpenRouter selected but OPENROUTER_API_KEY is not configured

AI-assisted analysis of mvanhorn/last30days-skill@c7460f6114 (2026-08-15). Data as JSON: /api/errors/65922828d819459b. Report an issue: GitHub.