chore: remove all Umbra references (Signetry naming only)

.github/workflows/reviewer.yml

@@ -1,6 +1,6 @@
 name: Reviewer
 
-# Advisory PR review (umbra-reviewer): surfaces architecture + security issues
+# Advisory PR review (signetry-reviewer): surfaces architecture + security issues
 # and posts one recommendation comment. Advisory only — it never merges and never
 # fails the PR. Changes to security-sensitive surfaces (workflows, packaging) are
 # escalated to a human. See https://github.com/Signetry/reviewer
@@ -22,9 +22,9 @@ jobs:
       - uses: actions/setup-python@v7
         with:
           python-version: "3.12"
-      - name: Install umbra-reviewer
+      - name: Install signetry-reviewer
         # source-available (All Rights Reserved); install from source, not PyPI.
-        run: pip install "umbra-reviewer @ git+https://github.com/Signetry/reviewer@v0.1.1"
+        run: pip install "signetry-reviewer @ git+https://github.com/Signetry/reviewer@v0.1.2"
       - name: Compute the PR diff
         env:
           BASE_SHA: ${{ github.event.pull_request.base.sha }}
@@ -40,7 +40,7 @@ jobs:
           PR: ${{ github.event.pull_request.number }}
         run: |
           set -uo pipefail
-          umbra-reviewer review \
+          signetry-reviewer review \
             --diff "$OUT/pr.diff" \
             --repo "$GITHUB_REPOSITORY" \
             --pr "$PR" \
@@ -58,7 +58,7 @@ jobs:
             const out = process.env.OUT || '';
             let body = 'Signetry Reviewer: no review was produced.';
             try { body = fs.readFileSync(path.join(out, 'comment.md'), 'utf8'); } catch (e) {}
-            const marker = '<!-- umbra-reviewer -->';
+            const marker = '<!-- signetry-reviewer -->';
             body = marker + '\n' + body;
             const { owner, repo } = context.repo;
             const issue_number = context.issue.number;

CHANGELOG.md

@@ -7,33 +7,33 @@ Until `1.0.0` the public API may change between minor versions.
 
 ### Changed — Signetry rename (breaking)
 
-- Renamed the distribution `umbra-eval` → `signetry-eval` and the import package
-  `umbra_eval` → `signetry_eval`. The console command is now `signetry-eval`.
-- All `UMBRA_*` environment variables are now `SIGNETRY_*`; the config directory
-  is `.signetry/` (was `.umbra/`). Product/brand prose updated to **Signetry**.
-- Core dependency renamed and pinned: `signetry-core @ git+https://github.com/Signetry/core@v0.6.0`
-  (was `umbra-core @ ...@v0.5.4`). Imports moved `umbra_core` → `signetry_core`.
+- Distribution `signetry-eval` and import package `signetry_eval`. The console
+  command is `signetry-eval`.
+- Environment variables use `SIGNETRY_*`; the config directory is `.signetry/`.
+  Product/brand prose updated to **Signetry**.
+- Core dependency pinned: `signetry-core @ git+https://github.com/Signetry/core@v0.6.0`
+  (was `signetry-core @ ...@v0.5.4`). Imports use `signetry_core`.
 - No backward-compatibility fallbacks are provided.
 
 ## [0.2.2] — 2026-07-30
 
 ### Changed
 
-- Install `umbra-core` from its **source repository** (`git+https://github.com/Signetry/core@v0.5.4`)
-  instead of PyPI — umbra-core is All Rights Reserved and no longer distributed on
+- Install `signetry-core` from its **source repository** (`git+https://github.com/Signetry/core@v0.5.4`)
+  instead of PyPI — signetry-core is All Rights Reserved and no longer distributed on
   PyPI. Enables `tool.hatch.metadata.allow-direct-references`.
 
 ## [0.2.1] — 2026-07-30
 
 ### Fixed
 
-- Dependency floor raised to `umbra-core>=0.5.0` (was `0.3.0`) so a fresh install
+- Dependency floor raised to `signetry-core>=0.5.0` (was `0.3.0`) so a fresh install
   always has the detection engine (`scan_repository`) the corpus benchmark needs.
 
 ### Docs
 
 - README now documents the **detection head-to-head benchmark** (52 cases, 7
-  languages; umbra-core 100% recall / 0 false positives vs Claude Opus 4.8 90%),
+  languages; signetry-core 100% recall / 0 false positives vs Claude Opus 4.8 90%),
   the `corpus` / `realrepo` commands, and the honest note that any false positive
   comes from the optional Semgrep layer, not the deterministic engine.
 
@@ -46,13 +46,13 @@ Until `1.0.0` the public API may change between minor versions.
   academic/CWE, crafted, hard (cross-file taint, framework sinks, true-negative
   traps), multilang, and cross-file-lang — with cited provenance per case and safe
   decoys for false-positive measurement.
-- `umbra-eval corpus` scores Umbra (live, via `umbra-core`) against competitor
+- `signetry-eval corpus` scores Signetry (live, via `signetry-core`) against competitor
   scanners (replayed from committed captures), reporting recall, false positives,
   and a by-language breakdown. `--min-recall` / `--max-fp` gate a regression;
   `--semgrep` enables the optional layer.
-- `umbra-eval benchmark` (14-vuln fixture head-to-head) and `umbra-eval realrepo`
+- `signetry-eval benchmark` (14-vuln fixture head-to-head) and `signetry-eval realrepo`
   (live scan of real vulnerable repos).
-- Committed head-to-head result: umbra-core **100% recall / 0 false positives** vs
+- Committed head-to-head result: signetry-core **100% recall / 0 false positives** vs
   claude-code-security-review (Claude Opus 4.8) 90% — deterministic, offline, free.
 
 ### CI
@@ -64,4 +64,4 @@ Until `1.0.0` the public API may change between minor versions.
 
 - Initial adversarial evaluation suite: ASR (ungoverned vs governed) + utility-
   under-defense across IPI, skill/MCP poisoning, and memory-injection scenarios,
-  run against the real `umbra-core` admission pipeline.
+  run against the real `signetry-core` admission pipeline.