agent: builder model: opus max_turns: 110 budget_usd: 25 —
peek(..)/not(..) fails at run) and #1050 (a guard handing an optional label to a by-value Node? parameter reaches rustc)Score: two defects y2b found in the yaml lane (c8d95984): nikaia check passes, the run does not. Nothing in packages/yaml needs them now (it uses a named rule peek(Dash) and t: ref Node?, both sound as written); they touch only the grammar checker, emitter and lowering, so this lane builds them before #1000 (y3). Two issues, two commits: #1049 first (small, probed), then #1050 (needs a reduction first). Read docs/builder-facts.md; y2b’s report docs/briefs/yaml/y2b-keys-directives-limits.md.report.md, its Found: (2) and (3).
Part II 10 (docs/specification/20-nikaia-advance.md, the element table): not(p) and peek(p) take a pattern p, and p q is a pattern. The checker accepts a sequence there; the run refuses it.
grammar G {
entry rule doc -> String = peek("-" "a") s:until(";") ";" -> { s }
}
fn main() {
let r = G::doc("-a;") catch {
println("error")
return
}
println(r)
}Today (develop 5817b28d, probed by the architect): nikaia check prints Checked main.nika; nikaia run stops with Error: Unexpected token, expected ). at the second element inside peek(..); the same with not("+" "a"). ("-" "a")? and text("-" "a") run. Wanted: -a for both. (A syntactic rule skips whitespace before each element, so a probe with a " " element after another inside a lowercase rule fails for that reason, not this one: keep the probe’s elements without spaces.)
To build: (1) find which parser refuses it at run (nikaia run parses the grammar a second time somewhere the checker does not, or the emitter writes the built-in’s argument as one element) and make it take a sequence inside peek(..) and not(..), as text(..) does; try every other one-pattern built-in of the table (until, bytes, span, count, dec[T], intern) the same way and fix those that refuse it; (2) test tests/compiler/tests/grammar_a_builtin_takes_a_sequence.nika + .stdout, failing on the unfixed binary. Fixes #1049.
packages/yaml/src/main.nika:567-568:
| k:FlowAny(ds, left) t:FlowTail(ds, left - 1)?
if(pair_fits(left, k, t), "nesting within the depth limit") -> { .. match t { .. } }with fn pair_fits(left: i64, k: ref Node, t: ref Node?) -> bool sync (main.nika:992) compiles. With t: Node?, nikaia check passes and rustc refuses the lowered guard closure: “closure only implements FnOnce”. A 20-line probe with a small struct did not reproduce it. The guard reads t by value and the action reads t again afterwards: a guard must lend what it reads, or the checker must refuse the move with an existing NK code; a rustc error is a defect either way.
To build: (1) reproduce: copy packages/yaml to a scratch directory (outside the checkout), change line 992 to t: Node?, NIKAIA_CACHE_LIMIT=2G target/fast/nikaia test --project .; read the generated Rust under the run’s target/nikaia/gen/; shrink (drop rules the failing rule does not call, then fields of Node) until a probe of a few dozen lines fails; name what makes it fail (a non-Copy field, the optional label, the label read again in the action, the sync marker). (2) Fix it where the guard’s call is checked or lowered (check_grammar.nika, emit_pattern.nika’s guard closure, crates/nikaia/src/contracts/keeps.rs if the guard’s call site counts as keeping, see 7a91e7d9 for the entry’s case), so a by-value parameter fed from a label in a guard is treated as an ordinary call’s argument that is read again afterwards. (3) Test tests/compiler/tests/grammar_a_guard_hands_an_optional_label_by_value.nika + .stdout (or .stderr if the checker rightly refuses), failing on the unfixed binary. packages/yaml keeps ref Node?. Fixes #1050.
If the reduction does not reproduce in 25 turns, commit #1049, leave #1050 open with a comment saying what was tried (gh api repos/nikaia-foundation/nikaia/issues/1050/comments), and report partial with Needs: #1050 reduction.
Both tests pass and fail on the unfixed binary (say so under Tests:); NIKAIA_CACHE_LIMIT=2G target/fast/nikaia test --both-settings --project tests/compiler > out.txt 2>&1 green with every grammar_ test in it; the package gate below green; each fix its own commit with its Fixes #N, pushed. Files you may change: crates/nikaia-std/src/tools/check_grammar.nika, emit_pattern.nika, emit_grammar.nika, the grammar parser the run uses (name it under Body:), their lowered twins (crates/nikaia-std/src/tools/lowered/*.rs, regenerate as docs/builder-facts.md says), crates/nikaia/src/contracts/keeps.rs, the two new tests with their expected output. Name under Changelog: what each change decided.
docs/builder-facts.md first. develop in the main checkout; git status at start: other agents’ uncommitted files are not yours, never touch or stage them.git add <files> only; commit after each step; git push origin develop fast-forward, no rebase (rejected: report Pushed: rejected, go on committing).cargo fmt, cargo clippy --all-targets -- -D warnings, scripts/pre-push-check.sh, and the gate of .claude/agents/builder.md (#852).CHANGELOG.md, badges; no new NK code. New tests in Nikaia only; never the full suite. No foreground sleep, no polling; scripts in files.std or the toolchain lacks stops the work: no code around it unless an ADR or Part I-III refuses it (name it). Else Result: blocked, Found: gap + probe, Needs: owner decision (CLAUDE.md, A missing form stops the work).NIKAIA_CACHE_LIMIT=2G; at most two clones with target/ in the machine; delete your own clone’s target/ and target/release/nikaia-scratch at the end; report GB freed.Pushed: line names the pushed sha. A report without a Pushed: line, or with a placeholder left in (TC, DISK), is not done and cuts the batch (k3a1 2026-10-29: Result: done, nothing committed).NIKAIA_CACHE_LIMIT=2G target/fast/nikaia test --project packages/yaml green, and cargo test --profile fast -p nikaia --test it -- fmt::every_nika_file (it formats every .nika under packages/). For a change only under packages/yaml/ this is the gate the line above asks for; a change under crates/ runs the gate of .claude/agents/builder.md (#852) as well. target/fast/nikaia is built by the session hook; if it is missing, cargo build --profile fast -p nikaia once.Changelog: none, unless the brief changes crates/.Files not to touch: anything outside packages/yaml/ and the brief’s own report, unless the brief names it; never docs/briefs/batch.txt or a brief of another lane.
max_turns: commit and push every finished step, then report Result: partial with what is left under Needs:. Uncommitted work at the end is lost work (y4 ran out at 110)./home/user/Nikaia/docs/briefs/yaml/y18-1049-1050-grammar-builtins-and-guard.md.report.md)Fields of docs/architect.md, Briefs and reports: Result: done|noop|partial|blocked (the bare word alone on that line: Result: done, never a summary after it, which cuts the batch, b869a 2026-10-27; Owner: is a question for the owner or none, never a role), Files, Commit, Body (optional, one paragraph for the commit message), Fixes, Pushed, Tests, Gate (green|red), Deviation, Found, Needs, Changelog, Owner, Disk freed, and for a move Rs2nika: (the kept N of M lines of rs2nika --score, then the three costliest hand edits; none for a task that moves no Rust, none (redesign) for a brief with Shape: redesign and no Convert: line). none where nothing, and then the bare word: Deviation: none (a note) cuts the batch (batch.nika compares exactly); a note goes into the commit message. Found: is only for a defect outside your change (it cuts the batch); a process event such as another session resetting or rewriting your files is no Found:, it goes under Needs: (m3a 2026-10-28); a converter draft that came out wrong is no Found:, it is one of the costliest edits under Rs2nika:, which the converter lane reads (e12-4b 2026-10-27); how you solved your own step goes into the commit message. Deviation: is only a step of the brief built otherwise than written or not built (a choice the brief left to you, such as keep what X needs and say so, is no deviation: it goes into the commit message; nor is building two steps in one commit when every step is built, e12-3a 2026-10-27, nor a step another commit on develop already built, shown by your test failing with that commit’s change reverted, b889 2026-10-27: name the commit in Body:, nor a test’s expected .stdout/.stderr taken from what the program really prints where the brief’s sketch of it was short or wrong, b934 2026-10-29: say so in Body:, nor a lowering that follows the ADR governing it where the brief’s expected lowering contradicts that ADR, b933 2026-10-29: name the ADR in Body:, nor a lowering other than the brief’s where rustc cannot take the brief’s, b1029 2026-10-31: say why in Body:, nor re-keying an allow-list entry (KNOWN) whose hash changed for the same program and the same reason, b-fixforward-sweep2 2026-10-28: say so in Body:, nor an existing test’s expected .stdout/.stderr updated because your change rightly changes the message it prints, b1012b 2026-10-31: name the test in Body:); how you verified (e.g. a test not run against the unfixed compiler) goes under Tests:. A noop says why under Tests: (what was probed), not under Deviation:. Final answer, one line: done <sha>, report <path> (or BLOCKED: ... / PUSH REJECTED ...).
Attribution: Co-Authored-By: Claude Opus 5.5 noreply@anthropic.com Claude-Session: https://claude.ai/code/session_01UVxJDveRM5qKZKyCqq6Y24
The first builder on this brief was stopped mid-work when its driver hit a time limit. #1049 (step 1) is built and committed (0c9f96a4, gate green, report y18-1049-1050-grammar-builtins-and-guard.md.step1.report.md). Build #1050 only, and copy step 1’s Changelog: line into your report beside your own. The unfinished #1050 work is saved outside the checkout: /tmp/claude-0/-home-user-Nikaia/e431aae9-4983-5166-8583-f9b3eb3b6f61/scratchpad/y18-step2-partial/ (tracked.diff for check_grammar/check_takes, plus the started test grammar_a_guard_hands_an_optional_label_by_value.*). Read it, take over what holds up, and check every piece of it. It has not passed any gate.