Repository navigation
feat(scripts): hermes_parity — parity-sync CLI (status/start/gates/bisect/ack/finish/clean) - #265
Merged
Merged
Conversation
…sect/ack/finish/clean) Automates the upstream/fork parity-sync workflow distilled from the 2026-07-10 sync (PR #255: 56 conflict files, 147 hunks, ~45 CI-surfaced reconciliation bugs). Stdlib-only package at scripts/hermes_parity/, run as 'python -m hermes_parity' via root shim. Surface: - status [--fail-behind N]: ahead/behind, merge-base, conflict summary - start: isolated worktree + frozen target + staged merge + conflict-bucket report (docs/sync/review/conflict-buckets.md) + atomic state file - gates [--stage|--fast|--resume|--strict]: 6-stage ladder (markers, imports, unbound-name AST lint, manifest+forkdelta, full tests, linux-only listing) with per-stage repro commands, gates.jsonl, tree-SHA invalidation, CI-owned reminder checklist (pinned-gitleaks version parse incl. fleet VER= form, hard-warns on parse failure) - bisect: merge-vs-baseline classification (REGRESSION/INHERITED/ UPSTREAM TEST/ORDER-POLLUTION/FLAKY), cached baseline worktree with SHA-mismatch re-pin, bounded parallelism, --from-file -, flake re-run - ack --reason: first-class reviewed-and-intentional clearance for fork-delta hard-fails (distinct from finish --force) - finish: all-green verification, 2-parent merge commit, push, generated PR body + gh command; --force requires --force-reason - clean: worktree removal with rollback echo Design reviewed: 4-model mixture-of-agents panel (v2), Momus pass-1 BLOCK (4 header/body seams, reconciled in v2.1), Momus pass-2 APPROVE-WITH-CHANGES (RC1-RC5, folded in v2.2 spec + this code): fork-delta trigger pinned to the merge-touched delete/rename subset with honest coverage split; ack path; gates signature reconciled; baseline cache invalidation; module name pinned hermes_parity. Manifest seeded with 5 fork features from the 2026-07-10 postmortem (docs/sync/fork-features.json). Verified: 16 unit tests green; live smoke against the real repo (start created a real 3-conflict worktree vs upstream b9b463f, buckets correct, markers gate exit 1 on real conflicts, ack persisted, clean removed worktree); py_compile clean; all subcommand --help OK.
Kyzcreig
enabled auto-merge
July 10, 2026 16:37
Kyzcreig
added a commit
that referenced
this pull request
Jul 26, 2026
…00 summary-role pin) (#431) * vendor(lcm): cherry-pick 9 upstream fixes from hermes-lcm (03b74f8 -> selected from 49e99a2) Upstream hermes-lcm went MIT (LICENSE added 2026-06-26, f7ae61f) — vendoring and cherry-picking now permitted with attribution. Tier-1 picks per /tmp/lcm-refresh-out/CHERRY-PICK-LIST.md, code-only (their tests/docs/changelog excluded; our vendored copy carries no tests dir — coverage rides tests/context_engine/): #263 preserve source lineage after long sessions #264 perf: aggregate DAG status stats #265 harden externalized payload durability #269 preserve raw session ownership across compression rollover #278 avoid payload integrity false positives from log examples #280 pin summary role to user after system anchor (Anthropic HTTP 400) <- highest value #285 make context engine deepcopy clone-safe (subagent spawn safety) #282 strip injected context before compaction (+) discard reasoning-only summaries (unclosed <think> = quality bug + prompt leak) All 9 verified clean-apply by the refresh-analysis worker on a simulated copy of our tree; re-applied here onto fork/main. * test(compaction): regenerate in-turn reconcile fixture for the vendored sanitizer The 9 LCM cherry-picks (3c1d61c) add one line to _sanitize_active_context_messages (upstream pick #282, strip injected context before compaction), which moves the fixture's sanitizer_source_sha1 provenance hash and reds test_fixture_sanitizer_provenance_current. Regenerated via the committed generator: python tests/agent/fixtures/gen_inturn_reconcile_fixture.py Diff is the provenance hash ONLY -- messages, compressed, true_kept_count and fresh_tail_count are byte-identical. The new line routes through _preserved_objective_context_content, which returns "" unless a row starts with the preserved-objective prefix, so it is a strict no-op on all 632 fixture rows (verified: 0 rows mutated) and the real sanitizer output is unchanged. Only one of the four hashed functions changed; the other three are byte-identical. Follow-up candidate (not this PR): this provenance test is a change-detector, which AGENTS.md discourages -- it should assert the sanitizer's behaviour (live sanitize(raw_tail) == committed comp tail) rather than pinning its source SHA. --------- Co-authored-by: Kyzcreig <9063726+Kyzcreig@users.noreply.github.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
hermes_parity— parity-sync CLIWhat
Stdlib-only internal CLI (
scripts/hermes_parity/,python -m hermes_parity) that automates the upstream/fork parity-sync workflow: isolated worktree setup, conflict bucketing, a 6-stage verification gate ladder, merge-vs-baseline test classification, fork-feature survival enforcement, and PR handoff.Distilled from the real 2026-07-10 sync (PR #255: 56 conflict files, 147 hunks, ~45 CI-surfaced reconciliation bugs across 4 CI rounds). Goal: next sync goes from a full day to ~1 hour of genuine decisions.
Command surface
status [--fail-behind N]startgates [--stage] [--fast] [--resume] [--strict]gates.jsonl, tree-SHA invalidationbisectack --reason <why> <paths>--forcenever becomes routine)finishgh pr createcommandcleanKey mechanisms
docs/sync/fork-features.json— the "must survive any merge" registry ({feature, tests, paths, why}); stage 4 runs exactly those tests. Seeded with 5 features from the 2026-07-10 postmortem..parity-state.json— atomic writes, tree-SHA gate invalidation, rollback SHAs, acks.finish:--mergenever squash; if BEHIND, merge fork/main in, never rebase.Review provenance
b9b463f3; markers gate exit 1; ack persisted; clean removed)Spec:
docs/sync/2026-07-10-hermes-parity-SPEC.md· Operator guide:docs/sync/README-hermes-parity.md