Move trusted tool preparation into arena

[vaibhav-i]

Summary

• move pinned Bubblewrap, Codex, Claude Code, and optional Turso installation policy into ax-arena/benchmark/scripts/prepare-trusted-tools.sh
• leave GitHub-only inputs, secret binding, permissions, environment review, and artifact actions in the workflow launcher
• preserve the existing credential-free preparation job and strengthen shell failure behavior

Validation

• npm test — 559 core tests and 132 arena tests passed on Node 22
• npm run typecheck
• npm run build
• npm run pack:check
• both npm dry-run packs
• npm run check:boundaries
• bash -n ax-arena/benchmark/scripts/prepare-trusted-tools.sh
• git diff --check
• independent security and behavior-parity audit: clean

Compatibility / rollback

Workflow inputs, tool pins, install destinations, smoke tests, credentials, and artifact flow are unchanged. Rollback is a single commit revert.

Next slice

Move arena-only environment template ownership out of the public ax-eval package without changing active runtime aliases.

[github-actions]

Normalized records diff

Overall is a macro-average of participating surface scores. Agents are never averaged together.

Overall by agent

Product	Agent	Base pass@1	Head pass@1	Δ	Base pass³	Head pass³	Surfaces
fixture-db	claude-code	92.9%	92.9%	+0.0 pp	78.6% (11/14)	78.6% (11/14)	2
fixture-db	codex	78.6%	78.6%	+0.0 pp	64.3% (9/14)	64.3% (9/14)	2

Surface cells

Product	Agent	Surface	Base pass@1	Head pass@1	Δ	Base pass³	Head pass³	Latency ms	Cost USD	Tokens	Harness version
fixture-db	claude-code	api	85.7%	85.7%	+0.0 pp	71.4% (5/7)	71.4% (5/7)	1100 → 1100	0.4200 → 0.4200	3350 → 3350	2.1.5 → 2.1.5
fixture-db	claude-code	cli	100.0%	100.0%	+0.0 pp	85.7% (6/7)	85.7% (6/7)	1300 → 1300	0.5100 → 0.5100	3720 → 3720	2.1.5 → 2.1.5
fixture-db	codex	api	85.7%	85.7%	+0.0 pp	71.4% (5/7)	71.4% (5/7)	1200 → 1200	— → —	3600 → 3600	0.121.0 → 0.121.0
fixture-db	codex	cli	71.4%	71.4%	+0.0 pp	57.1% (4/7)	57.1% (4/7)	1500 → 1500	— → —	4100 → 4100	0.121.0 → 0.121.0

Operational metrics are context only and never affect ranking.