feat(arg-frg): add Language Grades Matrix + grade extraction to tracker

Surfaces each language's grade against the five hyperpolymath
language-grade standards (CRG/TRG/ARG/FRG/RSR) as an at-a-glance matrix
in TOOLING-STATUS.adoc. The values are mirrored from each language's own
`spec/{ARG,FRG,TRG}-PROFILE.adoc` "Current X Grade:" line; the
language-status-tracker.jl script gains a regex-based grade extractor
keyed off the stable profile format.

Cross-axis invariants documented:
* ARG ≤ TRG always (a language can't be more adoptable than its toolchain)
* ARG-A requires FRG ≥ B (top-tier adoption needs foundations)

This is Phase 5 of the ARG/FRG language-maturation framework (per
[[session-2026-05-28-arg-frg-framework-design]] in operator memory). The
framework itself is documented in standards/adoption-readiness-grades/
(separate landing); the per-language profiles for AffineScript / Ephapax /
My-lang are in their respective repos. This commit is the consumer view —
the dashboard that reads them.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: hyperpolymath

TOOLING-STATUS.adoc

@@ -11,6 +11,64 @@ This monorepo contains 12 programming languages + 2 meta-projects. All 12 langua
 are registered in the BoJ Server `lang-mcp` cartridge with session management, type-checking,
 and evaluation via zig adapter + Zig FFI.
 
+== Language Grades Matrix
+
+Per-language grades against the five hyperpolymath language-grade standards:
+
+* *CRG* (Component Readiness Grades) — worst-of any released component
+* *TRG* (Toolchain Readiness Grades) — language toolchain (parser, type checker, codegen…)
+* *ARG* (Adoption Readiness Grades) — language as adoptable artefact
+* *FRG* (Foundations Readiness Grades) — mathematical foundations
+* *RSR* (Rhodium Standard Repositories) — repository hygiene compliance level
+
+All graders use `X | F | E | D | C | B | A`. The grade source of truth lives in each
+language's repo at `spec/{ARG,FRG,TRG}-PROFILE.adoc`; the values below are mirrored
+here for at-a-glance comparison and are refreshed by `language-status-tracker.jl`.
+
+Cross-axis invariants: ARG ≤ TRG always; ARG-A requires FRG ≥ B.
+
+[cols="<2,^1,^1,^1,^1,^1,<3", options="header"]
+|===
+| Language | CRG | TRG | ARG | FRG | RSR | Notes
+
+| **AffineScript**
+| TBD | TBD | D | E | FULL
+| ARG-D leaning; CLI gates programs. FRG-E pragmatic / X strict (no `formal/`).
+
+| **Ephapax**
+| TBD | TBD | E | D | FULL
+| Strong FRG narrative (preservation stated, layered redesign committed). Adoption surface sparse.
+
+| **My**
+| TBD | TBD | D-leaning-E | X | FULL
+| 11 crates, 4 dialects. PROOF-NEEDS.md honestly records template-ABI removal (FRG-X discipline).
+
+| **Eclexia** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Betlang** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Phronesis** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Wokelang** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Anvomidav** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Julia-the-viper** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Oblibeny** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Error-lang** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Kitchenspeak** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Tangle** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **Me-dialect** | TBD | TBD | TBD | TBD | TBD | Profile not yet authored.
+| **VCL-total (vcl-ut)** | TBD | TBD | TBD | TBD | TBD | Database/identity language; per ARG §1.1 tightening applies.
+| **KRL** | TBD | TBD | TBD | TBD | TBD | Database/resolution language; canonical READINESS.md template lives here.
+| **Quandledb** | TBD | TBD | TBD | TBD | TBD | Database language.
+| **TypeLL** | TBD | TBD | TBD | TBD | TBD | Type-theory foundation, feeds VCL-total levels.
+| **PanLL** | TBD | TBD | TBD | TBD | TBD | Type-theory foundation.
+|===
+
+Standards specs:
+
+* `standards/component-readiness-grades/`
+* `standards/toolchain-readiness-grades/`
+* `standards/adoption-readiness-grades/`
+* `standards/foundations-readiness-grades/`
+* `standards/rhodium-standard-repositories/`
+
 == Language Tooling Matrix
 
 [cols="<2,^1,^1,^1,^1,^1,^1,^1,^1,^1", options="header"]

language-status-tracker.jl

@@ -101,6 +101,66 @@ function language_type(status)
     end
 end
 
+# --- Grade extraction (ARG / FRG / TRG / CRG / RSR) -------------------------
+#
+# Each language's per-grade profile lives at `spec/{ARG,FRG,TRG}-PROFILE.adoc`
+# in the language's own repo. The "Current X Grade:" line is the source of
+# truth for the language's grade against that standard. We extract by regex
+# rather than parsing AsciiDoc, because the profile format is stable.
+
+struct LanguageGrades
+    arg::String
+    trg::String
+    frg::String
+    crg::String
+    rsr::String
+end
+
+const GRADE_LINE_RE = Dict(
+    :arg => r"Current ARG Grade\s*\|\s*\*?\*?([XFEDCBA][^*|]*)"i,
+    :trg => r"Current TRG Grade\s*\|\s*\*?\*?([XFEDCBA][^*|]*)"i,
+    :frg => r"Current FRG Grade\s*\|\s*\*?\*?([XFEDCBA][^*|]*)"i,
+    :crg => r"Current CRG Grade\s*\|\s*\*?\*?([XFEDCBA][^*|]*)"i,
+    :rsr => r"RSR Compliance\s*\|\s*([A-Z]+)"i,
+)
+
+function extract_grade(path::String, kind::Symbol)
+    isfile(path) || return "TBD"
+    try
+        text = read(path, String)
+        m = match(GRADE_LINE_RE[kind], text)
+        return m === nothing ? "TBD" : strip(m.captures[1])
+    catch
+        return "TBD"
+    end
+end
+
+function get_language_grades(name::String)
+    path = joinpath(REPOS_DIR, name)
+    isdir(path) || return LanguageGrades("—", "—", "—", "—", "—")
+
+    arg_file = joinpath(path, "spec", "ARG-PROFILE.adoc")
+    frg_file = joinpath(path, "spec", "FRG-PROFILE.adoc")
+    trg_file = joinpath(path, "spec", "TRG-PROFILE.adoc")
+
+    arg = extract_grade(arg_file, :arg)
+    trg = extract_grade(trg_file, :trg)
+    frg = extract_grade(frg_file, :frg)
+    # CRG and RSR may appear in any of the three profile files; first hit wins.
+    crg = "TBD"
+    rsr = "TBD"
+    for f in (arg_file, frg_file, trg_file)
+        if crg == "TBD"
+            crg = extract_grade(f, :crg)
+        end
+        if rsr == "TBD"
+            rsr = extract_grade(f, :rsr)
+        end
+    end
+
+    return LanguageGrades(arg, trg, frg, crg, rsr)
+end
+
 function status_emoji(status)
     if !status.exists
         return "❌"
@@ -159,6 +219,24 @@ println()
 println("=" ^ 80)
 println()
 
+# Grade matrix — sources from spec/{ARG,FRG,TRG}-PROFILE.adoc in each repo.
+# Cross-axis invariants: ARG ≤ TRG always; ARG-A requires FRG ≥ B.
+println("LANGUAGE GRADE MATRIX (CRG / TRG / ARG / FRG / RSR)")
+println("-" ^ 60)
+println("| Language | CRG | TRG | ARG | FRG | RSR |")
+println("|----------|-----|-----|-----|-----|-----|")
+for status in sort(statuses, by=s->s.name)
+    status.exists || continue
+    g = get_language_grades(status.name)
+    println("| $(status.name) | $(g.crg) | $(g.trg) | $(g.arg) | $(g.frg) | $(g.rsr) |")
+end
+println()
+println("Grade source of truth: each language repo's `spec/{ARG,FRG,TRG}-PROFILE.adoc`.")
+println("Missing/unparsed entries surface as TBD; '—' means repo not on disk.")
+println()
+println("=" ^ 80)
+println()
+
 # Production ready list
 println("🟢 PRODUCTION READY (40+ source files):")
 println("-" ^ 40)