fix(tri): gates preview asks the contexts the ruleset requires, parse-ratchet among them (Closes #5725) by dmitrii-f-t27 · Pull Request #5732 · gHashTag/t27

dmitrii-f-t27 · 2026-10-03T10:13:18Z

tri gates preview asked four contexts it called "the four that can block a merge" (check, check-now-freshness, validate, check-linked-issue), which was the ruleset of 2026-09-06. The ruleset's last edit (2026-09-19 15:06 UTC, 21 s after #4277 merged the parse ratchet) leaves validate, check-linked-issue and parse-ratchet (gh api repos/gHashTag/t27/rules/branches/master; classic protection is off: branches/master reports protection.enabled: false). So two rows that cannot block a merge decided the exit code (an empty branch always exited 1), and parse-ratchet was never asked.

What changes

The set is read, not written down. Every run asks the ruleset (required_contexts, the reader tri gates required already used). If it can't be read, the set comes from .github/required-contexts.txt, a ledger that tri gates required --write regenerates (it writes only when the set changes, so the date never moves by itself). Both commands print drift between the ledger and the ruleset. If neither can be read, the preview says the set is unknown, asks every reader anyway, and exits 1.
A required context with no reader prints UNAVAILABLE, instead of being left out. A reader is used only when exactly its own workflow posts the context. GitHub matches a required check by name, so a renamed job or a second workflow taking the name turns the row UNAVAILABLE.
parse-ratchet runs its job's own steps. The steps are read out of spec-parse-ratchet.yml on every run: --self-test, then cargo build --release -p t27c (the checker gets the executable cargo reports, so CARGO_TARGET_DIR can't hand it a stale binary), then check_specs_still_parse.py over base..HEAD. Exit codes are the checker's: 0 PASS, 1 FAIL (the row names the specs), 2 UNAVAILABLE. validate works the same way, so it now also runs the --self-check it used to skip. If a job's steps differ from what its reader runs, the row reads UNAVAILABLE. This takes the idea behind the_pattern_is_read_out_of_the_gate_that_enforces_it and applies it to whole jobs.
check and check-now-freshness print under "NOT REQUIRED BY THE RULESET" and no longer decide the exit code. The NOW-gate scratch control (test_now_gate_writes_nothing.py --tri) still sees the preview reach the NOW gate's pass.
The documents that copied the set:
- docs/BRANCH-PROTECTION.md now states the ruleset as read. It used to list five required workflows; two of them were.
- spec-parse-ratchet.yml moves into MERGE_CRITICAL; it had been excluded with the reason "not a required check".
- LOOP-RULES, verify.sh, the Makefile, the pre-push hook and two workflow headers now point at the ledger instead of naming "the four required checks".
- tri gates required goes from 6 claims / 4 hollow / 1 unclaimed to 5 / 2 / 0. The two left are the docs/now workflows that MERGE_CRITICAL still lists (see below).
Help text. tri gates --help showed required's description against preview and unmeasured's against tests; each now sits on its own variant.
Census. fetches moves gates.rs:3651 -> 3668 (fn unmeasured), re-blessed in the same commit. This is a line shift from the doc comments added above it; the population is unchanged.

Verified locally

cargo build -p tri --all-targets (no new warnings); cargo test -p tri: 834 passed, 0 failed (13 in gates::preview_tests, 10 of them new).
Mutation: 7 hand-written mutants of the new code, each killed by a named test over the full 13-test sample.
Probes in a throwaway worktree:
- a commit breaking specs/tools/tri/gates.t27 gives FAIL parse-ratchet ... specs/tools/tri/gates.t27 and exit 1;
- an extra job step, a removed self-test, or a renamed validate job each give UNAVAILABLE with the reason;
- an unreadable ruleset falls back to the ledger, and the reason is printed;
- a ledger with a line removed gives LEDGER DRIFT from both commands;
- with neither source readable: "THE REQUIRED SET COULD NOT BE READ" and exit 1.
tri skill check, tri census pin --gate, tri gates tests --gate, check_pr_branch_filters.py (and --self-test), test_now_gate_writes_nothing.py --tri, check_now_entry_shape.py, check_fix_carries_source.py: all pass. The shared .git/config hash is unchanged across runs.

Not decided here

Whether check and check-now-freshness should be required again is the owner's call. Only an admin can edit the ruleset. Until that's decided they stay in MERGE_CRITICAL, so their pull_request trigger stays unfiltered, and tri gates required reports them as claimed but not required.

🤖 Generated with Claude Code

…-ratchet among them (Closes #5725) `tri gates preview` asked four contexts it called "the four that can block a merge": check, check-now-freshness, validate, check-linked-issue -- the ruleset of 2026-09-06. Since the ruleset's last edit (2026-09-19 15:06 UTC, 21 s after #4277 merged the parse ratchet) it requires validate, check-linked-issue and parse-ratchet. So two rows that cannot block a merge decided the exit code (an empty branch always exited 1), and the one new required context was never asked. - The set is read from the ruleset on every run (`required_contexts`), else from .github/required-contexts.txt, a ledger `tri gates required --write` regenerates only when the set changes. Both commands print drift. With neither readable the preview asks every reader and exits 1. - A required context with no reader prints UNAVAILABLE instead of being left out. A reader is used only when exactly its own workflow posts the context. - parse-ratchet runs its job's own steps, read out of spec-parse-ratchet.yml on every run: --self-test, `cargo build --release -p t27c` (the executable cargo reports), check_specs_still_parse.py over base..HEAD. validate does the same with its two steps, so it now runs its --self-check too. A job whose steps differ from its reader's reads UNAVAILABLE. - check and check-now-freshness print under "NOT REQUIRED BY THE RULESET" and no longer decide the exit code. - docs/BRANCH-PROTECTION.md states the ruleset as read; it listed five required workflows, two of which were. spec-parse-ratchet.yml joins MERGE_CRITICAL (its exclusion reason was "not a required check"). LOOP-RULES, verify.sh, the Makefile, the pre-push hook and two workflow headers point at the ledger instead of naming "the four required checks". - `tri gates --help` showed required's description on preview and unmeasured's on tests; each now sits on its own variant. Census: `fetches` moves gates.rs:3651 -> 3668 (fn unmeasured). A line shift from the doc comments added to GatesCmd above it; the population is unchanged. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

github-actions · 2026-10-03T10:17:01Z

PR Dashboard

Generated at: 2026-10-03 10:16:55 UTC

specs/port/tools: SKIP_SUFFIX, USER_OPCODE and the jtag tests are t27 (#5735) #5736 specs/port/tools: SKIP_SUFFIX, USER_OPCODE and the jtag tests are t27 (#5735) | Created: 2026-10-03 | Updated: 2026-10-03 | 1 failing / 2 total
specs/demos: restore the jones_signature_throughput bench and make two broken blocks readable #5734 specs/demos: restore the jones_signature_throughput bench and make two broken blocks readable | Created: 2026-10-03 | Updated: 2026-10-03 | 29 failing / 33 total
fix(tri): gates preview asks the contexts the ruleset requires, parse-ratchet among them (Closes #5725) #5732 fix(tri): gates preview asks the contexts the ruleset requires, parse-ratchet among them (Closes #5725) | Created: 2026-10-03 | Updated: 2026-10-03 | 15 failing / 36 total
specs/fpga: test checks written as 'invariant <expr>;' become then/and assertions #5731 specs/fpga: test checks written as 'invariant ;' become then/and assertions | Created: 2026-10-03 | Updated: 2026-10-03 | 16 failing / 37 total
specs/port/fpga: ported FPGA tests are read again (#5728) #5729 specs/port/fpga: ported FPGA tests are read again (#5728) | Created: 2026-10-03 | Updated: 2026-10-03 | 2 failing / 32 total
specs/math: sacred_physics checks every member of its finite sets instead of a dropped prose tail #5727 specs/math: sacred_physics checks every member of its finite sets instead of a dropped prose tail | Created: 2026-10-03 | Updated: 2026-10-03 | 2 failing / 36 total
specs/numeric: GF12 invariants check concrete points instead of prose the parser drops #5724 specs/numeric: GF12 invariants check concrete points instead of prose the parser drops | Created: 2026-10-03 | Updated: 2026-10-03 | 2 failing / 37 total
spec(crm-story-reel): v20 -- the hook reel as a template, yellow captions, the widget call, RU+EN #5697 spec(crm-story-reel): v20 -- the hook reel as a template, yellow captions, the widget call, RU+EN | Created: 2026-10-03 | Updated: 2026-10-03 | 2 failing / 32 total
Port gHashTag/trinity:fpga/openxc7-synth/d_e21_test.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/d_e21 #5689 Port gHashTag/trinity:fpga/openxc7-synth/d_e21_test.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/d_e21 | Created: 2026-10-03 | Updated: 2026-10-03 | 4 failing / 31 total
spec(automation): kanban-card-chat v7 -- the bee's own log on a running card, live #5680 spec(automation): kanban-card-chat v7 -- the bee's own log on a running card, live | Created: 2026-10-03 | Updated: 2026-10-03 | 5 failing / 32 total
Port gHashTag/trinity:fpga/openxc7-synth/clk_test.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/clk_tes #5675 Port gHashTag/trinity:fpga/openxc7-synth/clk_test.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/clk_tes | Created: 2026-10-03 | Updated: 2026-10-03 | 4 failing / 31 total
docs(constitution): entrench Article CREDIT and seal it (v1.3) #5667 docs(constitution): entrench Article CREDIT and seal it (v1.3) | Created: 2026-10-03 | Updated: 2026-10-03 | 4 failing / 40 total
specs/port/fpga/verilog/ternary_mac_synth.t27: rewrite w_code_to_mult/product in t27 form (explicit return, element-wise #5664 specs/port/fpga/verilog/ternary_mac_synth.t27: rewrite w_code_to_mult/product in t27 form (explicit return, element-wise | Created: 2026-10-03 | Updated: 2026-10-03 | 3 failing / 31 total
Port gHashTag/trinity:fpga/openxc7-synth/symbiflow_blink.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/ #5663 Port gHashTag/trinity:fpga/openxc7-synth/symbiflow_blink.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/ | Created: 2026-10-03 | Updated: 2026-10-03 | 5 failing / 31 total
docs(igla): re-take the corpus figures on master, 1150 specs (Refs #5497) #5610 docs(igla): re-take the corpus figures on master, 1150 specs (Refs #5497) | Created: 2026-10-02 | Updated: 2026-10-02 | 1 failing / 34 total
feat(xilinx7): 7-series configuration packets as a .t27 spec (Closes #5608) #5609 feat(xilinx7): 7-series configuration packets as a .t27 spec (Closes #5608) | Created: 2026-10-02 | Updated: 2026-10-02 | 2 failing / 32 total
Port gHashTag/trinity:fpga/openxc7-synth/d_clk_direct.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/d_c #5595 Port gHashTag/trinity:fpga/openxc7-synth/d_clk_direct.v (Verilog, 1 module) to specs/port/trinity/fpga/openxc7-synth/d_c | Created: 2026-10-02 | Updated: 2026-10-02 | 6 failing / 31 total
The Queen's accept is the merge: dispatch.t27 says so, and its harness measures it #5452 The Queen's accept is the merge: dispatch.t27 says so, and its harness measures it | Created: 2026-10-01 | Updated: 2026-10-02 | 6 failing / 36 total
docs(tri): mark the Solana tri-mining program and depin.prove rewards superseded by mint-on-acceptance (Closes #5412) #5413 docs(tri): mark the Solana tri-mining program and depin.prove rewards superseded by mint-on-acceptance (Closes #5412) | Created: 2026-10-01 | Updated: 2026-10-02 | 6 failing / 32 total
Port gHashTag/BrowserOS:trios/agent-server/apps/server/src/lib/container/image-loader.ts (TypeScript, 2 functions) to sp #5320 Port gHashTag/BrowserOS:trios/agent-server/apps/server/src/lib/container/image-loader.ts (TypeScript, 2 functions) to sp | Created: 2026-09-30 | Updated: 2026-10-02 | 8 failing / 31 total
A "using" in prose became an import; and the per-spec delta check compared strings #5215 A "using" in prose became an import; and the per-spec delta check compared strings | Created: 2026-09-29 | Updated: 2026-10-02 | 12 failing / 41 total
Auto-loop 2026-09-29: silent-discard parser family + loop infrastructure #5084 Auto-loop 2026-09-29: silent-discard parser family + loop infrastructure | Created: 2026-09-28 | Updated: 2026-10-02 | 10 failing / 45 total
Carry wave-loop-898's stranded work: green lean 24/24 (fixes red lean CI), mcp registry spec, research corpus #5081 Carry wave-loop-898's stranded work: green lean 24/24 (fixes red lean CI), mcp registry spec, research corpus | Created: 2026-09-28 | Updated: 2026-10-02 | 10 failing / 36 total
chore: drop build artifacts tracked since ea15cd54c (#1399), ignore *.rlib #5078 chore: drop build artifacts tracked since ea15cd54c (#1399), ignore *.rlib | Created: 2026-09-28 | Updated: 2026-10-02 | 7 failing / 40 total
Port gHashTag/BrowserOS:trios/agent-server/apps/server/src/api/services/queen-brief-shape.ts (TypeScript, 2 functions) t #4911 Port gHashTag/BrowserOS:trios/agent-server/apps/server/src/api/services/queen-brief-shape.ts (TypeScript, 2 functions) t | Created: 2026-09-27 | Updated: 2026-10-02 | 11 failing / 31 total
Seals: 0 of 1072 specs verify, 928 with an unchanged spec_hash — gate it or remove it #4887 Seals: 0 of 1072 specs verify, 928 with an unchanged spec_hash — gate it or remove it | Created: 2026-09-27 | Updated: 2026-10-02 | 12 failing / 40 total
Wave Loop 473 — compiler-backend aggregate tail / live cold-POR CCLK sweep #4801 Wave Loop 473 — compiler-backend aggregate tail / live cold-POR CCLK sweep | Created: 2026-09-24 | Updated: 2026-10-02 | 11 failing / 31 total
Bare for statements: 2 loops recovered, and 235 of the 240 sites I cited were out of scope #4777 Bare for statements: 2 loops recovered, and 235 of the 240 sites I cited were out of scope | Created: 2026-09-24 | Updated: 2026-10-02 | 10 failing / 31 total
Generic type parameters dropped from the declaration and kept in the body: -23 errors #4745 Generic type parameters dropped from the declaration and kept in the body: -23 errors | Created: 2026-09-24 | Updated: 2026-10-02 | 12 failing / 31 total
Implement the 8 empty function bodies in specs/vsa/packed_vsa.t27 #4729 Implement the 8 empty function bodies in specs/vsa/packed_vsa.t27 | Created: 2026-09-24 | Updated: 2026-10-02 | 12 failing / 31 total
Wave 684: occlusion four deep, and the two files no list ever reached #4660 Wave 684: occlusion four deep, and the two files no list ever reached | Created: 2026-09-23 | Updated: 2026-10-02 | 10 failing / 31 total
Capture given/when/then and render them: 6540 clauses, and 169 specs share one unfilled placeholder #4642 Capture given/when/then and render them: 6540 clauses, and 169 specs share one unfilled placeholder | Created: 2026-09-23 | Updated: 2026-10-02 | 10 failing / 31 total
Implement part 2 of 2 of the empty bodies in specs/git/diff.t27 (7 functions) #4634 Implement part 2 of 2 of the empty bodies in specs/git/diff.t27 (7 functions) | Created: 2026-09-23 | Updated: 2026-10-02 | 10 failing / 31 total
Port part 1 of 2 of tools/check_withdrawn_live.py to specs/port/tools/check_withdrawn_live.t27 (8 functions) #4581 Port part 1 of 2 of tools/check_withdrawn_live.py to specs/port/tools/check_withdrawn_live.t27 (8 functions) | Created: 2026-09-22 | Updated: 2026-10-02 | 9 failing / 31 total
Port part 1 of 2 of tools/check_coq_in_build.py to specs/port/tools/check_coq_in_build.t27 (8 functions) #4578 Port part 1 of 2 of tools/check_coq_in_build.py to specs/port/tools/check_coq_in_build.t27 (8 functions) | Created: 2026-09-22 | Updated: 2026-10-02 | 8 failing / 31 total
Implement the 1 empty function body in specs/test_framework/verilog_bench_harness.t27 #4565 Implement the 1 empty function body in specs/test_framework/verilog_bench_harness.t27 | Created: 2026-09-21 | Updated: 2026-10-02 | 0 failing / 2 total
Implement part 2 of 2 of the empty bodies in specs/vm/jit_semantics.t27 (4 functions) #4563 Implement part 2 of 2 of the empty bodies in specs/vm/jit_semantics.t27 (4 functions) | Created: 2026-09-21 | Updated: 2026-10-02 | 8 failing / 31 total
Implement the 1 empty function body in specs/test_framework/graph_drift_detection.t27 #4536 Implement the 1 empty function body in specs/test_framework/graph_drift_detection.t27 | Created: 2026-09-21 | Updated: 2026-10-02 | 8 failing / 31 total
Nothing builds bindings/wasm-explorer, which is how the artifact it replaces drifted in the first place #4498 Nothing builds bindings/wasm-explorer, which is how the artifact it replaces drifted in the first place | Created: 2026-09-21 | Updated: 2026-10-02 | 6 failing / 30 total
Feed the swarm from CI, with fuel that does not run out #3978 Feed the swarm from CI, with fuel that does not run out | Created: 2026-09-17 | Updated: 2026-10-02 | 6 failing / 28 total
coq(CorePhi): replace proofs that were never checked by a compiler #3343 coq(CorePhi): replace proofs that were never checked by a compiler | Created: 2026-09-06 | Updated: 2026-10-02 | 5 failing / 32 total
Land the formal verification harness on master (it has never run) #2727 Land the formal verification harness on master (it has never run) | Created: 2026-08-27 | Updated: 2026-10-02 | 5 failing / 24 total

Summary

Status	Count
Total Open PRs	42
PRs with Failing Checks	37
PRs with All Checks Green	5
READY	1
FAILING	37
PENDING	0
NO CHECKS YET	0

These columns do not partition: 1 + 37 + 0 + 0 = 38, and there are 42 open PRs. A PR is being counted twice or not at all.

Seal Status

⚠️ STALE -- sha256(compiler.rs)=b45a356c2eb6 != 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 · 2026-10-03T10:18:41Z

📓 NotebookLM Notebook linked to this PR

Notebook ID: b83263109fb055dc
URL: https://notebooklm.google.com/notebook/b83263109fb055dc

This notebook contains session context, decisions, and artifacts for this work.

dmitrii-f-t27 requested a review from gHashTag as a code owner October 3, 2026 10:13

This was referenced Oct 3, 2026

specs/fpga: test checks written as 'invariant <expr>;' become then/and assertions #5731

Merged

specs/demos: restore the jones_signature_throughput bench and make two broken blocks readable #5734

Merged

This was referenced Oct 3, 2026

tri gates preview reads PASS on check-linked-issue for a reference the gate strips from the body #5744

Open

fix(tri): gates preview reads the PR body the way the issue gate does (Closes #5744) #5745

Open

Navigation Menu

fix(tri): gates preview asks the contexts the ruleset requires, parse-ratchet among them (Closes #5725) - #5732

fix(tri): gates preview asks the contexts the ruleset requires, parse-ratchet among them (Closes #5725)#5732
dmitrii-f-t27 wants to merge 1 commit into
masterfrom
fix/preview-required-contexts

dmitrii-f-t27 commented Oct 3, 2026

Uh oh!

github-actions Bot commented Oct 3, 2026

Uh oh!

github-actions Bot commented Oct 3, 2026

Uh oh!

This branch has not been deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant

Conversation

dmitrii-f-t27 commented Oct 3, 2026

What changes

Verified locally

Not decided here

Uh oh!

github-actions Bot commented Oct 3, 2026

PR Dashboard

Summary

Seal Status

Uh oh!

github-actions Bot commented Oct 3, 2026

Uh oh!

This branch has not been deployed

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

1 participant