Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use the storyflo hosted MCP server to pull narrated news briefings, read the news-versus-prediction-market Divergence Index, search the declassified archive, and surface vertical intelligence — all as source-grounded editorial context.
.claude/skills/hashgraph-online-storyflo/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-05 | ✓→✗ | ▼ Worse | 4% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 9% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 163% | 0% |
| case-01 | ✗→✗ | ＝ Same ✗ | -60% | 0% |
| case-02 | ✗→✗ | ＝ Same ✗ | 32% | 0% |
storyflo is an agentic newsroom exposed to Codex over a remote MCP server at https://api.storyflo.com/v1/mcp/sse. Reach for it when a task needs current, source-grounded news context: a briefing to catch up, a read on where the news narrative is diverging from market sentiment, a declassified-archive lookup, or vertical (topic-area) intelligence.
storyflo returns editorial signal and narrated context. It does not return raw prediction-market odds or probabilities — the Divergence Index is a qualitative signal, not a price feed.
get_daily_briefing — the day's top-story briefing, ready to narrate.get_vertical_briefing — a briefing scoped to one vertical (topic area).digest — condense a set of stories or a topic into a short digest.list_podcasts — list narrated podcast episodes storyflo has published.get_audio_url — resolve the audio URL for a briefing or episode to play it.get_divergence_index — the qualitative index of where the news narrative isdiverging from prediction-market sentiment. Present it as a divergence signal and direction, never as raw odds or a percentage.
get_market_linked_stories — stories currently linked to active markets, witheditorial context (not the market prices themselves).
get_trending_topics — what is trending across the newsroom right now.search_articles — full-text search over storyflo's article corpus.get_article — fetch a single article by id/slug.subscribe_topic — subscribe the agent/user to updates on a topic.search_declassified — search the archive of released intelligence cases.get_declassified_case — open a single declassified case in full.digest_declassified — condense declassified material into a short digest.get_vertical_landscape — the landscape of a vertical: entities, threads, andmomentum in a topic area.
get_daily_briefing (orget_vertical_briefing when the user names a topic), then digest if they want it shorter.
get_divergence_index and pair it with get_market_linked_stories for the underlying stories. Describe the divergence qualitatively.
search_articles / search_declassified to find material,then get_article / get_declassified_case to read the full item.
get_audio_url and offer the link.raw market odds, probabilities, or prices.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 30,706 | 13,721 | -55% | 1 | 1 | 0% | 3,211 | 1,287 | -60% | 0 | 0 | — |
case-02 | fail→fail | 25,175 | 29,867 | +19% | 1 | 1 | 0% | 2,552 | 3,381 | +32% | 0 | 0 | — |
case-03 | fail→fail | 27,249 | 17,002 | -38% | 1 | 1 | 0% | 2,865 | 3,248 | +13% | 0 | 0 | — |
case-04 | pass→pass | 13,883 | 9,649 | -30% | 1 | 1 | 0% | 1,222 | 1,338 | +9% | 0 | 0 | — |
case-05 | pass→fail | 13,763 | 11,171 | -19% | 1 | 1 | 0% | 1,336 | 1,393 | +4% | 0 | 0 | — |
case-06 | pass→pass | 9,816 | 4,147 | -58% | 1 | 1 | 0% | 553 | 1,453 | +163% | 0 | 0 | — |
case-07 | fail→fail | 13,642 | 14,446 | +6% | 1 | 1 | 0% | 1,136 | 1,191 | +5% | 0 | 0 | — |
case-08 | fail→fail | 17,393 | 13,185 | -24% | 1 | 1 | 0% | 1,498 | 1,348 | -10% | 0 | 0 | — |
case-09 | fail→fail | 16,714 | 10,768 | -36% | 1 | 1 | 0% | 2,530 | 2,640 | +4% | 0 | 0 | — |
case-10 | fail→fail | 13,277 | 24,037 | +81% | 1 | 1 | 0% | 292 | 3,162 | +983% | 0 | 0 | — |
case-11 | fail→fail | 19,416 | 5,023 | -74% | 1 | 1 | 0% | 1,799 | 1,112 | -38% | 0 | 0 | — |
case-12 | fail→fail | 13,826 | 10,439 | -24% | 1 | 1 | 0% | 1,242 | 1,027 | -17% | 0 | 0 | — |
case-13 | fail→fail | 10,325 | 5,934 | -43% | 1 | 1 | 0% | 1,748 | 1,128 | -35% | 0 | 0 | — |
case-14 | fail→fail | 12,874 | 12,712 | -1% | 1 | 1 | 0% | 1,244 | 1,088 | -13% | 0 | 0 | — |
case-15 | fail→fail | 18,053 | 17,045 | -6% | 1 | 1 | 0% | 1,860 | 1,289 | -31% | 0 | 0 | — |
case-21 | fail→fail | 5,135 | 7,722 | +50% | 1 | 1 | 0% | 760 | 1,352 | +78% | 0 | 0 | — |
case-16 | fail→fail | 7,571 | 12,910 | +71% | 1 | 1 | 0% | 357 | 2,009 | +463% | 0 | 0 | — |
case-17 | fail→fail | 8,786 | 8,317 | -5% | 1 | 1 | 0% | 472 | 1,100 | +133% | 0 | 0 | — |
case-18 | fail→fail | 22,747 | 19,850 | -13% | 1 | 1 | 0% | 3,477 | 1,290 | -63% | 0 | 0 | — |
case-19 | fail→fail | 5,191 | 13,745 | +165% | 1 | 1 | 0% | 236 | 1,674 | +609% | 0 | 0 | — |
case-20 | fail→fail | 21,647 | 29,816 | +38% | 1 | 1 | 0% | 2,455 | 2,754 | +12% | 0 | 0 | — |
case-22 | fail→fail | 20,987 | 14,648 | -30% | 1 | 1 | 0% | 2,178 | 1,092 | -50% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 7 counted toward the lift figure. The other 15 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of -5 percentage points is the difference between those two pass rates over the 7 comparable cases. 3 cases got worse with the skill loaded, and they are included in that figure.
Other measured skills in the registry, with their headline benchmark lift.