Address PR #568 feedback from @tmchow:
1. Framework coverage: add Nuxt, Astro, Remix, SvelteKit signatures to
detect-project-type.sh with recipe reference files for each.
2. Monorepo detection: when root detection finds zero matches, probe
subdirectories up to depth 3 for framework signatures. Output grammar
extended with type@dir and multiple:type@dir,type@dir forms.
3. Package-manager detection: new resolve-package-manager.sh script with
lockfile priority cascade (pnpm > yarn > bun > npm).
4. Port resolution: new resolve-port.sh with 8-probe cascade replacing
the inline bash in SKILL.md. Adds framework config parsing, .env
quote/comment stripping, and removes unreliable AGENTS.md grep.
Includes review-autofix: empty-string guard on framework config port
validation (P0), compact -p3000 Procfile syntax support (P1).
128 tests pass (72 new across 3 test files).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds /ce:polish-beta — the polish phase that runs AFTER /ce:review but
BEFORE merge. Starts a dev server from .claude/launch.json, generates a
user-testable checklist, lets the user mark items for fixes, and
dispatches polish sub-agents in parallel. Emits stacked-PR seeds for
oversized items and a replan seed when the whole batch is too big to
polish as one unit.
Five phases: input triage, branch/PR acquisition, entry gate, dev-server
lifecycle, checklist+size-gate+dispatch, envelope. Headless mode mirrors
ce:review's output envelope shape and terminates with `Polish complete`.
Cross-cutting changes:
- ce:review now writes metadata.json alongside findings so polish can
verify the artifact matches the current branch and HEAD.
- Tests pin the intentional resolve-base.sh duplication between
ce-review and ce-polish-beta (self-contained skill rule).
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Beta skills now use disable-model-invocation: true to prevent accidental
auto-triggering. Descriptions written as future stable descriptions with
[BETA] prefix for clean promotion. Updated solutions doc and AGENTS.md
promotion checklist to include removing the field.
New issue-intelligence-analyst agent that fetches GitHub issues via
gh CLI, clusters by root-cause themes, and returns structured analysis
with trend direction, confidence scores, and source mix. Designed for
both ce:ideate integration and standalone use.
Agent design:
- Priority-aware fetching with label scanning for focus targeting
- Truncated bodies (500 chars) in initial fetch to avoid N+1 calls
- Single gh call per fetch, no pipes or scripts (avoids permission spam)
- Built-in --jq for all field extraction and filtering
- Mandatory structured output with self-check checklist
- Accurate counts from actual data, not assumptions
- Closed issues as recurrence signal only, not standalone evidence
ce:ideate gains:
- Issue-tracker intent detection in Phase 0.2
- Conditional agent dispatch in Phase 1 (parallel with existing scans)
- Dynamic frame derivation from issue clusters in Phase 2
- Hybrid strategy: cluster-derived frames + default padding when < 4
- Resume awareness distinguishing issue vs non-issue ideation
- Numbered table format for rejection summary in ideation artifacts
Kieran Klaassen