Description grammar enumeration complete

The check is description_grammar_enumeration_complete. It is VERIFY-stage. It warns. It is off by default. It never adds a warning to a default audit until you enable it. It fires when three things are true together: an approved requirement's description enumerates a closed set of bracket-trailer tokens of the form [name: ...], the requirement traces to a parser, and that parser now recognizes more trailers than the description lists. The comparison is exact [name: token matching. It is not a fuzzy match. Internal-metadata trailers such as [reviewed:] stay out of scope. It does not write the YAML. It does not prove the Go.

This hop is the symmetric companion to description delta reviewed. That check only fires when a description is edited. A stale-but-unedited description that the parser quietly outgrew is invisible to it. This check compares the two sets structurally and fires on the gap, whether or not the wording was touched.

The two resolutions are alternatives, not steps. If the trailer is real, extend the description so it enumerates and explains the new [name:] shape, then re-review the requirement. If the omission is deliberate, that is a human risk-acceptance decision. Ask the human to record a waiver. Do not waive it yourself. Do not delete the parser trailer to silence the finding. Do not paste the bare token into the description with no explanation. That last move clears the set-difference and leaves the spec enumerating a trailer whose meaning it never states.

# SYS-REQ-1373  description enumerates ignore trailers [ki:] [assumption:]
# the traced parser later gained [category:]
# the description was never edited
# extra tests still green. the wording did not move
# proof config set project.checks.description_grammar_enumeration_complete.enabled true
# proof audit --check description_grammar_enumeration_complete
# [VERIFY] description_grammar_enumeration_complete
# 1 approved description omitted a trailer the parser now accepts
# SYS-REQ-1373  missing [category:]
# WARN
# proof help description_grammar_enumeration_complete

Read both sides before you edit anything. The description that claims a closed set, and the trailer the parser actually accepts. Enable it without committing the setting if you only want a targeted hop:

proof audit --check description_grammar_enumeration_complete \
  --set project.checks.description_grammar_enumeration_complete.enabled=true
proof spec show SYS-REQ-1373
proof req edit SYS-REQ-1373 \
  --description "<existing text, extended to enumerate and explain [category: ...]>"
proof audit --check description_grammar_enumeration_complete --verbose
proof help description_grammar_enumeration_complete

The teaching graph is still one approved requirement whose description lists a closed trailer set and whose traced parser now accepts one more. Close it by extending the description so it enumerates and explains the trailer, then re-review, or by asking a human to waive the hop if the omission is deliberate. Do not delete the parser trailer to make the sets match. Do not paste the bare token. The hop does not write that description for you. The hop does not waive itself.

# SYS-REQ-1373 description now enumerates [ki:] [assumption:] [category:]
# proof audit --check description_grammar_enumeration_complete
# 0 grammar-enumeration findings
# VERIFY may move on. pass sits on a list that matches the parser

The wording-moved hop stays on description delta reviewed. The rewrite hop stays on characterization testing and mirrors. Jama still authors. Proof vs Jama.

Warning when an approved description enumerates a closed [name:] set and the traced parser accepts a trailer the list omits. Off by default. Enable it, or a default audit never sees it. Default severity is warning, not error. It does not fail the merge by itself. Scope is the bounded ignore-trailer grammar the check documents, not every bracket in the tree. Exact [name: token matching only. No alias inference. Internal-metadata trailers such as [reviewed:] stay out of scope. Zero in-scope findings is a pass. A waived finding is a signed claim, not a proof that the omission is safe. The hop does not write the description. It does not edit the parser. It does not run proof waive for you. It does not prove the Go. A quiet hop is not a proof that Jama's shall matches the grammar, only that every in-scope closed list currently matches the parser or that nothing asked. We have not scored this floor against a frozen Jama pack or a VectorCAST corpus. The loss is named, not scored.

The wording-moved hop stays on description delta reviewed. The rewrite hop stays on characterization testing and mirrors. The engagement stays on software correctness audit. Jama still authors.

What is description grammar enumeration complete? Same question. Same URL.

Is this description delta reviewed? No. That hop reports a description-only edit on a FRETish-backed shall. This hop reports an unedited closed list the parser outgrew. See description delta reviewed.

Is this authored delta expected? No. That hop reports a missing no-authored-change stamp on a moved production file. This hop reports a missing trailer on a present description. See authored delta expected.

Is this no authored change surface reviewed? No. That hop reports a present refactor stamp over new public surface. This hop reports an omitted trailer. See no authored change surface reviewed.

Do extra tests clear an omitted trailer? No. A green suite is not a review of the grammar.

Does this finding fail the merge? No by default. The check keeps warning severity and stays off until you enable it.

Does a quiet hop prove the description matches the parser? No. The hop observes two token sets. It does not prove the Go.

Are internal trailers such as reviewed in scope? No. The hop only watches the documented exemption grammar.

Should an agent waive the omission? No. Accepting a spec that under-describes the grammar is a human decision.

Is this characterization testing? No. That page is rewrite-with-a-ledger. This hop is a closed list versus a parser. See characterization testing and mirrors.

Is Proof an LDRA alternative for the instrument? No. LDRA still owns the avionics toolchain. Proof vs LDRA.

Is Proof a Jama alternative for the shall? No. Jama still authors. Proof vs Jama.