/ln-42-dependency-upgrader

name: ln-42-dependency-upgrader
description: "Upgrades dependencies in reversible batches with version-specific compatibility and regression checks."

Dependency Upgrader

**Goal:** Upgrade dependencies in small, attributable batches. Preserve manifests, lockfiles, runtime support, and product behavior; do not treat a newer version as valuable without compatibility, security, or maintenance evidence.

**Execution contract:** Track every checklist item as `PENDING`, `PROVEN` with observed outcome evidence, `CLEARED` with evidence its condition is absent, or `UNPROVEN` with a gap. Reading, delegation, tool failure, zero exit status, and self-reported success are not proof. Reconcile after each section; before returning, resolve all `PENDING`, count only `PROVEN` and `CLEARED`, and apply verdict and approval rules to every gap. Preserve intent, scope, mode, and existing authorization. Complete authorized work before required external approval; ask only consequential unresolved choices. If nobody can answer, report the exact question and apply the skill's gap verdict without waiting or guessing. Cite any blocking instruction and unresolved boundary; never invent approval gates. Scale depth to material risk; preserve mandatory checks, dependencies, and safety order while choosing verification methods. Accept equivalent user or repository evidence without requiring another skill, named artifact, or full lifecycle. Preserve source requirement and decision IDs. Bind reused evidence to relevant versions, dirty changes, configuration, environment, and observation windows; invalidate only affected claims. Choose the primary workflow by outcome and explicit selection. Consume neighboring contracts and valid evidence without duplicating ownership or restarting audits. Split distinct deliverables by scope, not mandatory skill invocation. Broad independent assessments use audit workflows; focused domain reviews remain read-only. On continuation, reconcile task, authorization, state, and gaps. Return a compact continuation record for long work or update an already authorized artifact; read-only skills do not persist it. Distinguish artifact readiness, verified behavior, and external-action authority. Trace applicable requirements to their authoritative owner, planned or implemented change, acceptance evidence and status in the existing artifact or response; no new registry is required. Explicit constraints remain binding until an authorized deviation; risk determines priority, not permission to ignore them. Investigate the causal flow beyond initially named files. Within implementation authority, include structural repair needed for the outcome; minimize total complexity and risk, not changed lines. This does not authorize unrelated cleanup or mutation during review. Before retrieval, define the evidence question; read selectively without skipping applicable instructions. Use tools for mechanical processing; summarize bounded results, errors, coverage, and source paths, keeping full output outside context. Before technical decisions, establish applicability and actual versions; verify version-dependent assumptions against official documentation or specifications. Search and read primary pages for current external recommendations; snippets are not evidence. Version-matched bundled documentation or valid prior evidence suffices. Record decision, version, source, and limitations. Mark unavailable evidence UNVERIFIED; block only essential dependent work. Prefer native mechanisms, justify added tools or abstractions with evidence, and reuse the existing rule owner.

Tool Routing

| Need | Preferred tool | Use it when | Fallback | |---|---|---|---| | Package-manager detection | Manifests, lockfiles, workspace files, runtime files, and repository instructions | Always before choosing commands or update scope | Build and CI configuration | | Outdated and vulnerable packages | Native package-manager outdated and audit commands | The manager and registry are available | Official registry, vendor advisory, and lockfile inspection | | Breaking changes and support | Official release notes, migration guides, changelogs, advisories, and runtime support tables | Every consequential minor, major, replacement, or security update | Primary-source repository releases; otherwise mark `UNVERIFIED` | | Usage and blast radius | Language server or host-native code intelligence | An updated API, type, plugin, build tool, or runtime may affect consumers | Targeted import, symbol, configuration, and script search | | Safe version changes | Native package-manager commands | Updating manifests and generated lock state | Do not hand-edit lockfiles or emulate package resolution | | Verification | Repository-defined install, restore, build, lint, type, test, migration, and smoke commands | Before changes and after each batch | CI and script inspection with explicit unverified status | | Diff and rollback | Git status, diff, and isolated commits or worktree | Protecting user changes and reverting only the failed batch | Stop if the batch cannot be isolated safely |

Never publish packages, rotate credentials, deploy, or weaken audit and verification gates. Do not run lifecycle scripts from an untrusted package source without the environment's normal safeguards.

Evidence Rules

Version upgrades include container base images and runtimes when version transition is the requested outcome. Container engineering owns packaging redesign, image composition and process lifecycle. Reuse its existing contract without requiring a second workflow for an incidental packaging adjustment.

• Manifests declare constraints and lockfiles record resolved versions; install metadata proves the active environment when it differs. Registry "latest" does not override runtime or compatibility constraints.
• A vulnerability finding requires the affected version, advisory, reachability or exposure context, and a credible remediation.
• A breaking