Interface EvalArm

Reasoning budget for the run (claude --effort, e.g. "low" or an integer). Lives here beside model because it is part of the MEASUREMENT — it moves the output distribution, not the sample size — so it is hashed into the lock and the cache, and never read from an env var. Omit for the harness default.

Files written on top of the base fixture for this arm.

Tools to intercept for this arm (the tool-call spy). Each ToolIntercept is denied its real execution by an auto-wired PreToolUse hook — the model still emits the tool_use (so its arguments land in the Trace for toolWith / notTool), but the side effect (a paid API call, a git push, a paid subagent) never happens: the call is intercepted (prevented), NOT executed. This makes a real-model eval side-effect-free and safe (it does NOT cut the model-call cost); and because CC surfaces the denial as a blocked call, it's for asserting the agent's ATTEMPT, not for stubbing a tool to continue a multi-step flow. See src/tool-intercept.ts.

Model alias/id for THIS arm, overriding the eval-level model. A model comparison IS a harness A/B — arms: { sonnet: { model: "claude-sonnet-4-6" }, opus: { model: "claude-opus-4-8" } } — so model-as-an-arm answers "does my harness still hold on the cheaper tier / after a model upgrade?" through the same significance machinery, with no separate model-matrix runner. Omit to use the eval-level model. See research/eval-architecture.md (model strategy).

Path to a real plugin/repo to load for this arm (hooks + CLAUDE.md + skills). Lets an arm be "the whole plugin on" vs "off". See src/plugin-loader.ts.

Path to a plugin dir to install NATIVELY (claude --plugin-dir) for this arm, so its skills/commands/agents activate the real way — the real model can trigger a skill by its description (vs. plugin, which materializes a file subset that does not register skills). Point at a COMPLETE plugin. Lets an arm be "skill installed" vs "off" to measure real activation. Provide this OR skillsDir, not both.

.claude/settings.json (hooks/permissions) for this arm; omit for none.

A directory of LOOSE skills (<skillsDir>/<name>/SKILL.md, e.g. a repo's .claude/skills) to install for this arm. vigiles packages them into a throwaway --plugin-dir (manifest + skills/<name>/, each skill dir copied whole) for the run and removes it afterward — the one-liner for repo-local skills that aren't a published plugin. The skills install under the namespace vigiles-loose-skills, so a skill() check / skillResolved matches vigiles-loose-skills:<name> (the report's namespace says so). Provide this OR pluginDir, not both.