diff --git a/.grok/skills/unsga3-oracle/SKILL.md b/.grok/skills/unsga3-oracle/SKILL.md
index f14f8a1..28b6fc0 100644
--- a/.grok/skills/unsga3-oracle/SKILL.md
+++ b/.grok/skills/unsga3-oracle/SKILL.md
@@ -9,7 +9,7 @@ description: >
# Unsga3 oracle
-Repo root: Unsga3 (not Torquon). Confirm `Unsga3.slnx` / `tools/OracleCompare` exist before running.
+Repo root: Unsga3. Confirm `Unsga3.slnx` / `tools/OracleCompare` exist before running.
## Defaults (protocol)
diff --git a/CHANGELOG.md b/CHANGELOG.md
index e4c35fe..a9efa1e 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -11,11 +11,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/).
- Optional **TypeSafe / Jev System One** pass (`tools/typesafe-pareto`) for Score + Choice judgments over a small Pareto candidate sample. Additive semantic layer — does not replace NSGA-III / U-NSGA-III objectives. Live calls only when `TYPESAFE_API_KEY` is set; CI uses mocks. Metrics append to `metrics/typesafe-runs.jsonl`.
+### Changed
+
+- Docs and XML comments describe `initialPopulation` and hybrid loops in generic terms (domain-adapter warm-start / grid-seed). No product-repo names.
+
## [0.1.3] — 2026-08-10
### Added
-- Optional **`initialPopulation`** on `Unsga3Algorithm.Run` (grid-seed / warm-start for domain adapters; Torquon dogfood)
+- Optional **`initialPopulation`** on `Unsga3Algorithm.Run` (grid-seed / warm-start for domain adapters)
- Multi-seed Wilcoxon / Mann–Whitney oracle harness (`tools/oracle/run_multiseed_wilcoxon.py`)
- Community files: CONTRIBUTING, SECURITY, CODE_OF_CONDUCT, ROADMAP, CITATION.cff, issue templates
diff --git a/docs/OPERATOR-BOUNDS.md b/docs/OPERATOR-BOUNDS.md
index 7ae2b6d..4055fe5 100644
--- a/docs/OPERATOR-BOUNDS.md
+++ b/docs/OPERATOR-BOUNDS.md
@@ -9,4 +9,4 @@ That repository works on the still-open completely-bounded analogue and multi-op
- No code or data dependency is required by Unsga3.
- Do not open PRs or write into `crouzeix-extensions` from this repository (and vice versa).
-When Unsga3 is used in hybrid loops with physics-based local search or matrix-function evaluations (e.g. inside Torquon-GB or Hygra pipelines), the sharp scalar factor 2 can tighten residual and approximation bounds on non-normal operators.
+When Unsga3 is used in hybrid loops with physics-based local search or matrix-function evaluations, the sharp scalar factor 2 can tighten residual and approximation bounds on non-normal operators.
diff --git a/src/Unsga3/Algorithm/Unsga3Algorithm.cs b/src/Unsga3/Algorithm/Unsga3Algorithm.cs
index cafb638..985c225 100644
--- a/src/Unsga3/Algorithm/Unsga3Algorithm.cs
+++ b/src/Unsga3/Algorithm/Unsga3Algorithm.cs
@@ -83,8 +83,9 @@ public OptimizationResult Run(IProblem problem, int maxGenerations) =>
///
/// Optional seed individuals (decision variables only; re-evaluated). Used to inject a known
- /// feasible set (e.g. Torquon grid oracle). When null or empty, a uniform random init is used.
- /// Extra seeds beyond are truncated; fewer are pad-filled randomly.
+ /// feasible set (e.g. a domain-adapter warm-start or grid seed). When null or empty, a uniform
+ /// random init is used. Extra seeds beyond are truncated; fewer
+ /// are pad-filled randomly.
///
public OptimizationResult Run(
IProblem problem,