git:20260811.e13e6a2 to git:20260813.0a7e4f3

1 added, 1 removed. Audit A to A.

# Benchmark Instructions

- Use Fire for every Python CLI. Add a study to an existing provider runner when it shares transport or isolation; keep stage-specific contracts and scorers in focused modules rather than creating a second launcher.

- Interactive A/B/C result rows are a CLI contract: persist plain rows to a stage run log when that stage has one, and always route terminal output through the existing shared Rich arm renderer. Do not add direct `print()` paths for arm rows; redirected output must remain ANSI-free. Add a focused renderer-forwarding regression for every new stage or rescore path.

- - Never run paid models. Give the user an exact command with a freshly resolved `--paid-approval` SHA-256, then analyze only the artifact they provide.

+ - Never run paid models. Give the user the exact command emitted by a fresh dry run with its 16-character `--paid-approval` token; retain the complete SHA-256 in benchmark provenance, then analyze only the artifact they provide.

- Treat task suites, manifests, frozen repositories, indexes, and input snapshots as immutable benchmark coordinates. Regenerate generated manifests after contract or consumer changes; do not edit result artifacts.

- Convert benchmark findings into generic production contracts; never copy benchmark task IDs, target repositories, prompt wording, expected answers, or task-specific source/symbol examples into shipped plugins, Skills, templates, or user-facing docs.

- Keep A/B/C arms symmetric except for the documented treatment supplement. State Codemap's static-graph boundary: use it for compact symbol/dependency/importer/caller facts, not runtime validation, test execution, or edits.

- Treat A_plain versus C_strict as the decision-grade comparison. B_auto is an optional-use canary: if it costs more than A_plain, recommend the installed integration rather than treating that as a failure of the strict treatment.

- Executable tasks require benchmark-owned disposable worktrees, canonical diff capture, a second clean scoring worktree, ordinary patch application, independent behavior oracle, and verified cleanup. `--recount` is diagnostic-only.

- Run focused `pytest` for behavior, then invoke the exact pre-commit hooks for changed files: `pre-commit run ruff-check --files <python-files>`, `pre-commit run ruff-format --files <python-files>`, and `pre-commit run mdformat --files <markdown-files>`. For a changed runner execute its relevant `--dry-run` and scope-resolution command; no-model checks may be run by Codex.