Skip to content

spec(t27os): window petal and domain names equal the owner table (Closes #14004) - #14005

Merged
gHashTag merged 9 commits into
masterfrom
claude/petal-names-14004
Nov 15, 2026
Merged

gHashTag merged 9 commits into
masterfrom
claude/petal-names-14004

Conversation

@gHashTag

Copy link
Copy Markdown
Collaborator

Closes #14004. Refs #13895.

What changed (.t27 only)

  • AN_SLUGS equals the owner table of 2026-11-15; canonical ids renamed (@memory -> @actors, @knowledge -> @memory, @logs -> @telemetry, @extension -> @extensions, @yield -> @delivery); agent_network DOMAIN_FILE follows.
  • WORLD_NAMES (petal labels, tooltips, header), INSTRUMENT_LABEL (header, AGENTS line, home rows) are the upper-case slugs. AGENT_LETTERS, PETAL_LETTERS, LP_LETTERS and petal_letter_byte show "#" (byte 35) as the 27th letter.
  • identity_table.t27: NAME_TAB, DOMAIN_TAB, ARCH_TAB, REG_TAB, LAYER_TAB deleted; name_of reads AN_CODES, domain_of reads AN_SLUGS, archetype/register/layer are empty. agent_parts TOOL_PAGES and t27os_shell SH_DOMAIN_TITLES deleted (they read the sources). sh_domain_purpose is empty.
  • glass_window.t27, minimal: draw_identity draws a label and value only when the value is non-empty (rows ARCHETYPE, REGISTER, LAYER vanish); tip_addr comment and test now expect "0x#" (length 3, byte 35); two comments name AGENTS.
  • New test every_window_name_equals_the_owner_table in t27os_shell.t27 pins WORLD_NAMES, INSTRUMENT_LABEL and AGENT_LETTERS to AN_CODES/AN_SLUGS; agent_registry test verify_domain_names_equal_the_owner_table asserts all 27 cells.
  • Kept: AN_PURPOSES, an_domain_purpose, an_domain_layer, an_domain_register in agent_registry.t27 (no importer outside the file; marked NOT DRAWN).

Disagreements found (shown -> wanted)

  • AN_SLUGS K knowledge, L logs, M memory, X extension, Y yield -> memory, telemetry, actors, extensions, delivery
  • agent_network DOMAIN_FILE -> same five slugs
  • WORLD_NAMES (ARCHITECTURE, BUILD, COMPILER, MIGRATION ... DOCS, SECURITY) -> upper-case slugs
  • PETAL_LETTERS / petal_letter_byte / AGENT_LETTERS / LP_LETTERS "Ti" (T, 84) -> "#" (35)
  • INSTRUMENT_LABEL ("ARCHITECTURE ADR SOUL" ...) -> upper-case slugs; TOOL_NAMES "t27ti" -> "t27#"
  • identity_table NAME_TAB ("Alpha" ...), DOMAIN_TAB ("Architecture / ADR / SOUL" ...) -> AN_CODES/AN_SLUGS; ARCH/REG/LAYER tables -> nothing drawn
  • SH_DOMAIN_TITLES "T27TI" -> "T27#"; sh_domain_purpose returned a purpose -> empty
  • draw_identity drew labels for empty rows -> guarded

Not changed (follow-up)

specs/agents/*.t27 (LETTER_NAME "Ti", Greek names, purposes; no longer read by the window), specs/t27os/queen_dash.t27 QD_PIN_AGENTS "Ti Y", homes.t27 prose "Ti Security / Reserve", domain_pulse.t27 domain_of. The "#" glyph in the window font was not checked visually.

Not run

Tests, seals, gen, cargo, screenshot and visual regression were not run (hook refuses them; CI runs them). Only t27b build --macho specs/t27os/glass_window.t27 was run: it compiles. Seals of the touched specs are stale and need a reseal.

T27L: {"issue":14004,"pr":"draft","tests":"not_run","visual_regression":"not_run","compile_check":"t27b --macho glass_window ok"}

🤖 Generated with Claude Code

… source

AN_SLUGS equals the owner table; WORLD_NAMES, INSTRUMENT_LABEL, AGENT_LETTERS and the
identity names read or are pinned to AN_CODES/AN_SLUGS; the 27th letter is #; no purpose,
layer or register is drawn. Duplicate tables deleted. Tests not run locally.

Closes #14004
Refs #13895

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 17:00:07 UTC

Summary

Status Count
Total Open PRs 49
PRs with Failing Checks 44
PRs with All Checks Green 5
READY 1
FAILING 44
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 44 + 0 + 0 = 45, and there are 49 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…reak the identity import cycle (Refs #14004)

identity_table no longer imports agent_registry (cycle through dash and kanban); identity and glass_window read AN_CODES and AN_SLUGS directly; dash, gw_title and live_data PETALS expectations follow the owner table; shell name test pinned.

Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 17:11:20 UTC

Summary

Status Count
Total Open PRs 49
PRs with Failing Checks 45
PRs with All Checks Green 4
READY 1
FAILING 45
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 45 + 0 + 0 = 46, and there are 49 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 17:25:57 UTC

Summary

Status Count
Total Open PRs 48
PRs with Failing Checks 44
PRs with All Checks Green 4
READY 1
FAILING 44
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 44 + 0 + 0 = 45, and there are 48 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 17:38:42 UTC

Summary

Status Count
Total Open PRs 48
PRs with Failing Checks 44
PRs with All Checks Green 4
READY 1
FAILING 44
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 44 + 0 + 0 = 45, and there are 48 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@gHashTag

Copy link
Copy Markdown
Collaborator Author

Before / after still frames for the petal-name change.

Before: master 32c616f (binary sha256 a0052a56...). After: this branch at 26b78ac (binary sha256 9bfa22cb...); the later commits on this branch change seal files only. Both built with t27b build --macho + ld.

Each case: launch with T27OS_PAGE, T27OS_THEME, T27OS_STILL=1, read T27OS_WINDOW <n>, wait 10 s, screencapture -x -o -l <n> every 3 s until two consecutive frames have the same sha256 (at most 6). No key or click was sent.

case before sha256 (12) / bytes after sha256 (12) / bytes tries before / after changed
home dark 9c6768eeaeeb / 204662 9c6768eeaeeb / 204662 1 / 2 no
home light ee93a3320cf7 / 153862 2ab1468e9665 / 181133 1 / 1 yes
C dark d64f65922026 / 942857 9073814ede98 / 927637 2 / 3 yes
C light bac6989c57ea / 843687 bf00c6883811 / 927832 1 / 6 yes
F dark 8dec234f0a6c / 315092 2b5d40172ef2 / 312642 1 / 1 yes
F light 51773ac818c1 / 310127 684fc681354c / 309698 1 / 2 yes
Q dark 5b2e359673b2 / 420322 2ae4d641be8e / 422728 1 / 3 yes
Q light b0a93d4dfef7 / 395678 8feba85b529d / 389595 2 / 1 yes

Hashes were recomputed by the orchestrator from the 16 settled files.

What this does and does not show:

  • 7 of 8 cases differ between the builds, as a rename should; home dark is byte-identical.
  • It is a hash comparison, not a pixel diff: it cannot tell a renamed label from any other change. No tool that runs on the bench today compares PNG files, and no human or model has looked at the pictures.
  • Frames keep changing for several seconds under T27OS_STILL (t27os window: T27OS_STILL frames are not still - live pane content reaches the frame #14008). C light after settled only on the 6th capture, with one failed capture before it, so that row is weak.
  • Home light differs while home dark does not; not explained.
  • The PNG files are local (target/vr-14004/before, target/vr-14004/after), not uploaded.

T27L: {"issue":"14004","milestone":"unassigned","component":"petal and domain names","artifact":"specs/t27os (PR #14005)","screenshot":"local target/vr-14004 (16 settled files)","visual_regression":"not_run","accessibility":"not_applicable","tests":"pass","model":"sonnet","orchestrator":"claude_code","cost_tier":"paid_exception","status":"draft","blockers":["no pixel diff tool on the bench","T27OS_STILL frames not deterministic (#14008)"]}

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 17:55:51 UTC

Summary

Status Count
Total Open PRs 46
PRs with Failing Checks 43
PRs with All Checks Green 3
READY 1
FAILING 43
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 43 + 0 + 0 = 44, and there are 46 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…4468 (Refs #14004)

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 18:07:30 UTC

Summary

Status Count
Total Open PRs 46
PRs with Failing Checks 42
PRs with All Checks Green 4
READY 1
FAILING 42
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 42 + 0 + 0 = 43, and there are 46 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

…s their tables (Refs #14004)

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 18:33:15 UTC

Summary

Status Count
Total Open PRs 46
PRs with Failing Checks 42
PRs with All Checks Green 4
READY 1
FAILING 42
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 42 + 0 + 0 = 43, and there are 46 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

Co-Authored-By: Claude Fable 5.2 <noreply@anthropic.com>
@github-actions

Copy link
Copy Markdown
Contributor

PR Dashboard

Generated at: 2026-11-15 18:45:18 UTC

Summary

Status Count
Total Open PRs 46
PRs with Failing Checks 42
PRs with All Checks Green 4
READY 1
FAILING 42
PENDING 0
NO CHECKS YET 0

These columns do not partition: 1 + 42 + 0 + 0 = 43, and there are 46 open PRs. A PR is being counted twice or not at all.

Seal Status

  • ⚠️ STALE -- sha256(compiler.rs)=29c50ada25c9 != manifest seal=87e5cbd3ad94.
    The committed NMSE numbers were certified against an older compiler.rs.
    Run scripts/reseal-check.sh locally for the two-step reseal command (advisory; not a merge gate).

@gHashTag

Copy link
Copy Markdown
Collaborator Author

Gate reading on ca9cfd4 before merge.

@gHashTag
gHashTag marked this pull request as ready for review November 15, 2026 18:58
@gHashTag
gHashTag merged commit 25bb529 into master Nov 15, 2026
35 of 38 checks passed
@gHashTag
gHashTag deleted the claude/petal-names-14004 branch November 15, 2026 18:58
gHashTag added a commit that referenced this pull request Nov 16, 2026
…tal (Refs #13960, Refs #13957)

The petal-naming wave (#14005-era) introduced the 2-char petal name
'UX'; the extraction assert in every_petal_is_labelled_with_the_home_of_
its_agent required n >= 3 and broke on it (t27b: assert line 1232,
csv_field returns n=2 for 'UX'). The rename wave's own CI never ran
this test to conclusion on master (Spec Guards FAILURE on master
predates this fix).

Fix: relax the extraction assert to n >= 2 — the check still validates
that the petal name was extracted with its real length; 'UX' is a
legitimate petal name. test-report: 27/27 pass (was 26/1 FAIL). Seal
re-saved against the fixed spec; --verify all hashes MATCH.

Unblocks: the agent_parts reseal -> #13962's spec-guards -> the
keystone PR. Also documents the root cause of master's own Spec Guards
FAILURE (the same test on master's tree).
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

spec(t27os): window petal and domain names equal the owner table, one source (Refs #13895)

1 participant