/ln-41-surgical-change-implementer

name: ln-41-surgical-change-implementer
description: "Implements one scoped product feature or fix using existing contracts; not upgrades, performance tuning or establishing cross-cutting technical foundations."

Surgical Change Implementer

**Goal:** Deliver one approved product-code change through the smallest complete solution that satisfies the business outcome. Remove superseded code and avoid speculative abstraction, duplicate mechanisms, and custom infrastructure already provided by the repository or platform.

**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 capability | Fallback | |---|---|---| | Scope and repository policy | User request, linked task or plan, repository instructions, Git status, and focused diff | State bounded assumptions; stop if a choice would materially change product intent | | Runtime ownership and impact | Language server, host-native code intelligence, traces, schemas, routes, and configuration | Narrow symbol and text search followed by direct reads of definitions and consumers | | Existing capability | Repository code, manifests, lockfiles, platform APIs, and installed dependency source | Current official documentation and primary sources; mark uncertain claims `UNVERIFIED` | | Safe implementation | Focused editor, native formatter, package manager, and generation commands | Minimal manual patch that preserves user changes and generated-file ownership | | Verification | Repository-defined build, lint, type, test, smoke, and runtime checks | Smallest reproducible check that proves the protected contract; disclose coverage gaps | | External semantics | Official documentation, specifications, security advisories, registries, and upstream source | Primary-source technical material; use secondary expert guidance only for tradeoffs |

Do not add unrelated repository-wide cleanup, redesign, dependency work, or performance tuning. Structural changes required to fix the owning cause belong to the approved outcome, including internal interfaces and all affected consumers; an initial file list is not an architectural boundary unless explicitly constrained. Do not modify external systems, persisted data, public contracts, user experience, or unrelated dirty files without explicit authority.

Workflow Boundary

A bounded product feature or defect stays in this workflow even when it touches logging, configuration or access checks. A request whose primary outcome is establishing a cross-cutting technical contract belongs to the corresponding foundation capability. Apply existing project contracts here; do not establish a competing policy or require another skill for an incidental change.

An approved durable audit-ledger feature is product implementation: consume its requirements for records, access, retention, integrity