agent: builder model: opus max_turns: 80 budget_usd: 16 —
fs::Path maps from TrustedMap to HashMap (and the gate that missed it)Blocks: grammar_run (the GR move, b070adb9, reverted in 0fbe3d6b; it re-enters the main batch once this is on develop)
Score: compiler bug, a silent regression of the lowering (ADR-010 D5/D9: the maps lose the fast hasher) that keeps a self-hosting move out, and a gate that let a red CI through as green. Every new tools file with a host trait may set it off again (K3b, P5c carry a guard for it).
Facts: Where things live, Traps already met, Commands
Adding crates/nikaia-std/src/tools/grammar_run.nika (b070adb9) makes lower-std write lowered/project_members.rs with collections::HashMap where it wrote collections::TrustedMap (seen, here, renames_in’s parameter, at_of; project_members.nika :118, :124, :184, :238, :282), although project_members.nika did not change. Not a cache effect (fresh NIKAIA_LOWER_CACHE_SALT). The suite was green because gate.nika --all runs lower-std first and rewrites the drifted file; CI was red.
Probe (architect re-ran it on origin/develop 524c5286, 2026-10-31 18:40: seen, here, renames_in’s here turn HashMap): git show b070adb9:crates/nikaia-std/src/tools/grammar_run.nika > crates/nikaia-std/src/tools/grammar_run.nika (the lowered.rs hunk of b070adb9 no longer applies; lower-std rewrites the index itself), then NIKAIA_LOWER_CACHE_SALT=f$RANDOM NIKAIA_SYSROOT=$PWD/crates target/fast/nikaia lower-std and git diff --stat crates/nikaia-std/src/tools/lowered/project_members.rs. Restore only the files you applied, by name.
crates/nikaia-std/src/tools/trust.nika trusted_maps :282-290 puts the last :: segment of every library function with Provenance::Untrusted into env.sources. GrammarRunHost::manifest (a body-less host method, so untrusted) puts the bare word manifest there; in project_members_packages_of the parameter manifest (manifest.paths.iter()) then reads as an untrusted source and poisons the class of seen/here. Check where env.sources is matched (tm_deps_of, tm_call_deps): a source must be a call of that function (its full path, or a method call on a receiver of that type), never a variable or parameter that happens to share the name. If the trace is wrong, find the rule that demotes the maps; if the demotion is right by D9, stop and say why (Needs: architect decision).
Tests:.trust.nika (and its lowered twin). Test tests/compiler/tests/trusted_map_name_not_a_source.nika + .stdout (or a package test if one module cannot show it): a body-less trait method fn manifest(ref self) -> String with untrusted provenance and, in another function, a parameter or let named manifest feeding a collections::HashMap[fs::Path, String] whose keys are trusted; the lowering must keep TrustedMap (assert it the way the existing trusted_map* / hasher* tests in tests/compiler/ do: grep -rln TrustedMap tests/compiler/tests | head). Commit trust: an untrusted function's name is no source where a variable carries it (Part #1060).scripts/gate.nika (gate_lower_std :108, the fixpoint :121-127) check first that the committed lowered files are what this compiler lowers: after the first lower-std, git diff --exit-code --stat crates/nikaia-std/src/tools/lowered crates/nikaia-std/src/lowered* against the tested commit, and fail the step (red: lowered drift: <files>) on a difference when the commit under test did not change that .nika (git diff --name-only HEAD^ HEAD); a builder’s own uncommitted re-lowering in --cheap mode stays allowed. Update the gate’s dry-run test if one covers the step. Commit gate: a lowered file that drifts from what this compiler lowers is red (Part #1060).Part: #1060, not Fixes: step 4 (the GR re-land) goes into the main lane after this.
The new test fails on the unfixed compiler and passes through nikaia test --project tests/compiler --both-settings (by name), with the tests whose names contain trusted, hasher or map (run by name); with grammar_run.nika applied, lower-std leaves project_members.rs as committed (say so under Tests:, then remove the applied files by name); lower-std fixpoint clean; build, fmt, clippy, pre-push-check.sh; committed locally. Changelog: one line. Stop about 15 turns before max_turns, commit what is finished and report partial.
An edit is needed outside trust.nika, its lowered twin, scripts/gate.nika, the gate’s test and the new test; an error in another agent’s file.
Not yours, never stage: every file git status shows modified or untracked at your start (lowered/project_members.rs is committed as TrustedMap since 84fbfb99; with the probe applied, restore it, lowered.rs, grammar_run.nika and lowered/grammar_run.rs by name), and every docs/briefs/* file except your report.
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 (scripts/commit.nika -- REPORT does it); commit locally after each step and never push: the driver runs the complete suite and pushes (.claude/agents/builder.md, Cheap checks only; owner, 2026-10-31).cargo build --profile fast -p nikaia, lower-std and its fixpoint when a tools/*.nika changed, cargo fmt, cargo clippy --all-targets -- -D warnings, scripts/pre-push-check.sh, the tests the brief names and the ones your change touches. No gate.nika.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: architect 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.Result: partial with what you ruled out, the probe, and the suspected file and function (.claude/agents/builder.md).target/fast/nikaia run scripts/commit.nika -- REPORT, which writes Pushed: local <sha> into it. 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)./home/user/Nikaia/docs/briefs/bugs/b15-1060-trusted-map-name-collision.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_01GFeMqxRbtUMiGVyDMxsvEq