agent: builder model: sonnet max_turns: 90 budget_usd: 15 —
String? in tools/*.nika (NK1125, NK1104)Score: bugs before moving (CLAUDE.md, docs/priorities.md); a wrong refusal every move into tools/*.nika meets (M3b found it), and the message names a type the author never wrote. One file area (the check), apart from the fix-forward’s tools/fmt.nika.
Found by M3b (093923c, 2026-10-29). In a tools/*.nika file (compiled by nikaia lower-std), a mutable local initialised from a text literal and then changed in place is refused:
pub fn modules_probe_a() -> String {
let mut out = "a"
out.push_str("b")
return out
}nikaia lower-std refuses it with NK1125 (A String? may be missing) and NK1104 (returning String?). The same lines in a user program run with nikaia run are accepted. Writing the type (let mut out: String = "a") gets past it; that is the house spelling (445 uses), but the refusal is wrong: nothing here can be missing.
Probe: (1) add the function to any crates/nikaia-std/src/tools/*.nika and run target/fast/nikaia lower-std: NK1125 + NK1104; (2) the same function plus fn main() { println(modules_probe_a()) } run with nikaia run: prints ab.
lower-std’s check of one module: crates/nikaia/src/sysroot.rs:281 lower_std_module, which calls crate::check::check_against(..) with its own settings (about :302) and refuse_assumes_in_std; compare those settings with what nikaia run passes (the std/tools mode, the text tiers: tools/text_tiers.nika, a literal’s tier under the tools profile).crates/nikaia-std/src/tools/check_refusals.nika:154 (doc at :114, reaches_through_a_plain_value); NK1104 beside it.tests/compiler/tests/a_struct_field_without_its_null_arm_stays_nullable.*, crates/nikaia/tests/it/nullable.rs (where.nika).String? under lower-std and not under nikaia run, and fix it in the checker (or the tools-profile typing it uses) so both accept the probe. Test: tests/compiler/tests/a_text_literal_local_changed_in_place_is_text.nika with the probe (.stdout ab). Also put the probe into a scratch copy of a tools/*.nika file, run lower-std, see it pass, and restore the file by name (git checkout -- <that file> and its lowered .rs); say so under Tests:. Run lower-std once more on the clean tree: no lowered file but those your fix changes.Fixes #943.The new test passes (target/fast/nikaia test --project tests/compiler filtered to it), the NK1125/NK1104 tests above pass (it -- nullable and their tests/compiler programs), the gate of builder.md (a change to check/ or tools/check_*.nika runs nikaia test --both-settings --project tests/compiler), clippy, cargo fmt, scripts/pre-push-check.sh. Changelog: one sentence.
The difference is a deliberate rule of the tools profile written in Part I-III or an ADR (then Owner: names it and the choice), or the fix needs a file another agent has uncommitted (git status first: tools/view_changes.nika and crates/rs2nika/convert/nikaia.* were another session’s at the time of writing).
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.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./home/user/Nikaia/docs/briefs/b943-literal-local-push.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 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:); 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_01GFeMqxRbtUMiGVyDMxsvEq