Same harness, same rules: Argon used the code graph in 12/17 sessions, Opus only 4/17
rseroter · x · 2026-11-13
Developer davenporter built ast-graph, a code-graph plugin for Antigravity, and analyzed 32 sessions (5,426 steps, Nov 3–10) to compare how Claude Opus 5.5 and Gemini 4 Argon behave under identical harness and AGENTS.md rules.
- Argon used the code graph in 12 of 17 sessions (70.6%), issuing multiple queries in 11; it paid the turn to read each tool's MCP schema and followed the contract
- Opus used it in only 4 of 17 sessions (23.5%), defaulting to grep and sed in bash; it used the graph heavily only when a repo wired it into package.json, and reverted to grep even after AGENTS.md added one line of call syntax
- Plugin hooks kept the index fresh on every edit (hundreds of sync runs), so the graph was always available
The takeaway: models have different default habits even under the same rules — not necessarily right or wrong, but a real difference worth accounting for in agent workflows.
More from coding & agent
- Claude Code reads 40% of all developer docs, Mintlify data shows — handotdev · 2026-11-13
- Student's AI agent defended his school's network from a cyberattack after he left for college — heyneighbor · 2026-11-13
- Threedog.js: open-source Three.js engine lets Claude Code build, playtest and debug its own games — Gullible_Ranger1402 · 2026-11-13
- SGLang Summit agenda: LinkedIn serves ads model at 10K+ QPS under 80ms — DavidBennett__ · 2026-11-13
- Dev builds an OP-1-inspired skill for agentic coding, posted on GitHub Next — narphorium · 2026-11-13
- vork: open-source headless 3D/VFX engine for AI agents, renders black hole bending light in 18 min — ravisparikh · 2026-11-13