tracker: graph-2026-11-simplify — sequant 3.0 cut to the measured shape: D-1 removals, a halt on a repeated AC_NOT_MET, the ready gate's claim dropped, chain and multi-issue marked experimental, production-reaching Evidence handed to a person, the downstream pin, the release

Open
  • Is something wrong?
#1,645 0 comments 0 reactions 0 assignees View on GitHub

Maintainers usually reply within 1 day

Nobody has claimed this yet.

Assessment

Difficulty
5/5
Estimated time
Over a week
Newbie friendliness
15/100
Issue type
Feature
Clarity
Mostly clear
Activity status
Active
Tech stack
github, typescript
Domain
devtools

Research direction

Start with docs/internal/graph-2026-11-simplify/PROJECT-DISCUSSION.md and issue-graph.md, then inspect the linked nodes #1501, #1523 and #1602 and the D-1 clusters #1647, #1648 and #1649. The issue describes a multi-node release graph, with acceptance checks for the directory, label, node count and removal of placeholder edges. Done means the directory is on main and all thirteen labeled nodes have real edges.

Written by the indexing model from the issue text.

Description

graph-2026-11-simplify

Epic: graph-2026-11-simplify Tier: mechanical Execution: runner

What this graph is

The owner adopted the evidenced shape on 2026-11-03: across nine experiments and 172 scored runs, sequant matched a native Claude Code recipe on recall and on verified acceptance criteria, at 1.9 to 4.6 times its estimated cost; what it adds is the issue as input, spec's AC extraction and file list, an independent per-AC check, the per-AC table, a verdict that refuses, one fix pass on the gaps the verdict names, and a person at the merge. This graph moves main toward that shape within the I-1 scope (src/ and bin/; the plugin freeze that first set it lifted 2026-11-06, D-7 amendment) and ships it as 3.0. Founding doc: docs/internal/graph-2026-11-simplify/PROJECT-DISCUSSION.md (§N below). The measured build is tagged research-baseline-2026-11-03 (31c6f8bdc).

Decisions (owner, 2026-11-03; founding doc §7)

D-1: F-10 removes the model ladder and effort escalation (model-ladder.ts, effort-escalation.ts and their flags), --stacked, --reflect and /reflect (the skill text after the freeze lifts), --full-qa, and /test's label auto-add (#1631). The /test skill stays.
D-2: keep maxIterations 3 and halt when AC_NOT_MET repeats (the #1162 rule); the ready gate keeps its loop and drops its second-look claim from the docs.
D-3: --chain and multi-issue runs are marked experimental, with a warning that names the measured success rates; the code stays.
D-4: pre-register #1068 now for qa's sections on fixture #34, with the cost estimated before any run; exec and fullsolve later.
D-5: hand-label 50 of the 538 pre-tool blocks before any hook change.
D-6: a new graph, graph-2026-11-simplify, whose founding doc cites the synthesis and links the nodes that already exist in #1501, #1523 and #1602 rather than moving them.
D-7: not yet; src/ and bin/ work and #1068 go first under the freeze, and the lift stays with the plugin directory review.
D-8: an Evidence line that names an env file, a production CLI or a live URL becomes a pending row with the command shown, run by a person (as #1630 proposes for CI).

D-7 amendment (2026-11-06): the plugin freeze is lifted; the directory refused main under a Verified-organizations rule, so no review holds the plugin. I-1 stays as this graph's scope boundary; OQ-7 answered 2026-11-06: none join; they run as their own PRs after 3.0.

Nodes

See docs/internal/graph-2026-11-simplify/issue-graph.md for the DAG, waves and the placeholder map. Linked, not moved (D-6): #1501's F-10 (#1647, #1648 and #1649 are its D-1 clusters), #1523 (#1526, #1521, #1529, #1622), #1602 (#1605, #1606).

Acceptance Criteria

  • AC-1: The graph directory is on main. Evidence: git ls-tree -r --name-only origin/main -- docs/internal/graph-2026-11-simplify | grep -c '\.md$' → ≥ 16
  • AC-2: The graph label exists. Evidence: gh label list --search graph-2026-11-simplify --json name --jq 'map(select(.name=="graph-2026-11-simplify"))|length' → 1
  • AC-3: Every node carries the label. Evidence: gh issue list --label graph-2026-11-simplify --state all --limit 50 --json number --jq length → 13
  • AC-4: No filed body keeps a placeholder edge. Evidence: gh issue list --label graph-2026-11-simplify --state all --limit 50 --json body --jq '[.[].body|select(test("#S-[0-9]")|not)]|length' → 13 (bodies with no placeholder left)

Non-Goals

  • Moving #1501's, #1523's or #1602's nodes (D-6).
  • Any decision beyond D-1..D-8; a new one is a D-row in §7 of the founding doc.

Done when

The directory is on main, the thirteen nodes are filed with the label and real edges, and D-1..D-8 stand on this issue.

Dominant language
TypeScript
Stars
1
Forks
1
Avg merge
11h 3m
Merged PRs (30d)
84

Getting set up

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

More from sequant-io/sequant

All issues in sequant-io/sequant

Similar issues

More TypeScript issues

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.