# Changelog

## Score

- CAI 59 → 59 (+0.3)
- Rubric changed (rubric-2026.10.5 → rubric-2026.11.2) — scores are not directly comparable.

## Lenses

- Code Health 55 → 55 (+0.0)
- Architecture 89 → 87 (-1.6)
- Maturity 65 → 65 (+0.0)
- Readiness 59 → 59 (+0.0)
- Security 78 → 73 (-4.7)
- Accessibility 58 → 60 (+2.3)
- Performance 71 → 71 (+0.0)

## Resolved (9)

- Ambiguous 'bad' vs 'empty' distinction. `remove_bad` and `remove_empty` are distinct, but `remove_bad` is vague. Does 'bad' mean nil, empty strings, or invalid objects? Without seeing the implementation, the naming `remove_bad` is less descriptive than `remove_empty`. If `remove_bad` includes `remove_empty` logic, it's redundant. If they are distinct, `bad` is a poor name.
- Disclosure policy has no reporting contact
- Documentation: no installation or build instructions (docs/doc/top-level-namespace.html)
- Documentation: no project overview (docs/doc/index.html)
- Documentation: no project overview (docs/doc/top-level-namespace.html)
- Duplicate functionality across namespaces. `Doing.String` and `Doing.Completion.StringUtils` both expose `ltrunc` and `ltrunc!`. Additionally, `Doing.Completion.StringUtils` exposes `short_desc` which is not present in `Doing.String` (though `Doing.String` has `short_desc` in the list? No, `Doing.String` has `short_desc` in the list provided: `Method: String.short_desc()`). Wait, looking closely: `Doing.String` has `short_desc()`. `Doing.Completion.StringUtils` also has `short_desc()`. This is a clear duplication of utility methods between a core String extension and a Completion-specific StringUtils class.
- Inconsistent bang (!) usage semantics. In Ruby, `!` usually indicates mutation. `deep_freeze` and `deep_freeze!` likely both freeze, but one might be recursive vs shallow or have different side effects. Similarly for `thaw`. However, `deep_thaw` and `deep_thaw!` are present. If `deep_freeze` is non-mutating (returns new frozen hash) and `deep_freeze!` is mutating, this is consistent. But if both mutate, the `!` is redundant. The same applies to `thaw`. Without implementation, the presence of both suggests potential confusion if the distinction isn't strictly 'mutating vs returning new'.
- Inconsistent validation method naming. Most types use `good?()` for validation, but `Doing.String` uses `valid_id?()`. This suggests `valid_id?` might be specific to ID validation, but given the presence of `good?` on other String-like or general types, it creates confusion about whether `valid_id?` is a general validator or a specific one. If it is general, it should be `valid?` or `good?`. If it is specific, the name is misleadingly broad compared to `good?`.
- Potential overlap in configuration retrieval. `fetch` and `value_for_key` both retrieve values from configuration. `fetch` takes a path and default, while `value_for_key` takes a keypath. If `value_for_key` does not support defaults or handles resolution differently, it's a separate concern. However, the naming `fetch` vs `value_for_key` is inconsistent with standard Ruby conventions (`fetch` usually implies raising or defaulting, `[]` or `dig` is more common for lookup). If `value_for_key` is the primary accessor, `fetch` might be redundant or confusingly named.

## New (2)

- Disclosure policy has no reporting contact (SECURITY.md)
- Documentation: no build, run or usage example (README.md)

## Architecture

- Unchanged — 0 containers · 1 contexts · 0 edges
