Convert Claude Code or Codex into the most intelligent Deep Research Agent. Collect, search, and synthesize web research into a persistent, searchable wiki that builds on itself. Hosted API + MCP: hyperresearch.ai
Python agentsagentskillscitationsclaude-codecodexcodex-plugincodex-skilldeep-researchdeep-research-agentdeepresearchmcp-serverresearchresearch-agentresearch-tools
docs: benchmark chart — truncate y-axis at 30 for readability (#5)
Jordan Gibbs725edb5bdfe0
diff --git a/assets/_generate_benchmark.py b/assets/_generate_benchmark.pyindex f4049e2..e562ced 100644--- a/assets/_generate_benchmark.py+++ b/assets/_generate_benchmark.py@@ -114,15 +114,17 @@ for i, tick in enumerate(ax.get_xticklabels()):tick.set_fontweight("bold")tick.set_fontsize(13)-# Y-axis+# Y-axis — truncated at 30 so differentiation in the 45-60 range reads+# clearly. Scores in this regime are compressed; 0-100 full scale would+# flatten everything into identical-looking bars.ax.set_ylabel(- "RACE overall score (0–100, higher = better)",+ "RACE overall score (axis truncated at 30; max = 100)",color=dim,fontsize=11,labelpad=10,)ax.tick_params(axis="y", colors=dim, labelsize=10)-ax.set_ylim(0, max(scores) * 1.18) # headroom for star + labels+ax.set_ylim(30, max(scores) * 1.12) # headroom for star + labels# Gridlinesax.yaxis.grid(True, color=grid, linewidth=0.8, zorder=1)diff --git a/assets/benchmark.png b/assets/benchmark.pngindex 451e9dc..3aaaaf3 100644Binary files a/assets/benchmark.png and b/assets/benchmark.png differ