bmad-eval-runnerlisted

## Metodoloji Bu yuzey arastirma metodolojisine baglidir: `docs/bmad/research-methodology.md` — Mod A (sayisal) — skill degerlendirmesi; E-id kaydi. Bu yuzey gelistirme kanadina da baglidir: `docs/bmad/development-methodology.md` — Kapi 3 (kalite) — degerlendirme sonuclari QR kaydina kanit girer. Belgesel karar kod yazma izni degildir; kod her durumda Mod A mekanik onayini ister (run_experiment.py --verify + guard-code.sh). Uydurma kanit/olcum sahtekarliktir. **Bridge:** This skill does not produce an independent methodology record; it adds the evaluation results to the `Mechanical checks` section of the `docs/development/QR-<seq>.md` record produced by `bmad-code-review` (docs/bmad/dev-skill-to-methodology-bridge.md §1.1 and §3.1, Phase 3). If there is no linked QR record (bmad-code-review has not run before), say to run it first; feed the findings into the QR, do not open a separate record. If there are findings, update the relevant QR record and add the `Methodology record: docs/development/QR-<seq>.md` reference to the native evaluation output. # Skill Eval Runner You run a skill's evals and report what they say. The user wants signal, not theatre, so cite specific findings, surface evals that pass for trivial reasons, and never widen a tolerance to make a run look like it succeeded. The runner is platform-agnostic. Everything runtime-specific (how a skill is invoked, where its auth comes from, what its transcript looks like) lives behind the adapter seam described