Change or review Lvke MCP tool, Resource, schema, response-envelope, and migration contracts. Use whenever public tool interfaces, compact schemas, handlers, or dependent Skills change. This product has no frontend or HTTP application contract.
Implement and review Lvke MCP stdio servers, domain services, jobs, idempotency, persistence, schemas, and Resources. Use for MCP handlers, runtime behavior, storage, or tool-contract changes. This product has no authentication, tenant, role, RBAC, or permission-management layer.
Recover Lvke work from locks, provider failures, stale objects, proposal conflicts, finance inconsistency, or interrupted jobs. Use when a governed workflow is blocked, cancelled, stale, or resumable.
Orchestrate complete Lvke feasibility-study work across evidence, planning, finance, report, review, and release. Use for end-to-end feasibility delivery or when deciding the next governed domain stage.
Run the Lvke local verification and golden-sample matrix when remote CI is unavailable. Use before claiming code, MCP, report, finance, or desktop changes are complete.
Run live conversational acceptance for the 14 Lvke MCP services after code freeze and one restart. Use for MCP full testing, compressed-tool migration validation, or post-restart acceptance.
Draft, assemble, revise, validate, and export Lvke feasibility reports and DOCX deliverables. Use for any report chapter, document artifact, report revision, or formal report-generation task.
Run and recover Lvke source discovery and deep research. Use for policy, market, industry, technology, risk, comparable-project, or multi-source research tasks.
Review Lvke deliverables for evidence, consistency, numerics, format, policy, risk, decision support, and knowledge quality. Use for findings, remediation, deterministic rescoring, retest, comparison, or review export. This workflow has no authentication, role, permission, or security-signoff step.
Govern Lvke source import, security scan, parsing, evidence extraction, benchmarking, and source qualification. Use for project files, public snapshots, locators, evidence packs, or source quality work.
Coordinate Lvke MCP services, workspace navigation, precedent-first drafting, and multi-stage tool calls. Use when a task spans multiple Lvke domains or needs a resumable call sequence.
Plan and model Lvke urban rail transit (metro, light rail, tram) feasibility work. Use for rail ridership and clearing fare, alignment and station siting, per-kilometre investment intensity, car-kilometre operating cost, rolling stock renewal, and operating staffing. Do NOT use the generic public-service route for…
Checklist for changing Lvke MCP tool and Resource contracts. Use when editing tool names, input/output schemas, error codes, idempotency, Resources, finance runs, source files, deep research, reports, or migration mappings.
Job and idempotency rules for the local Lvke MCP runtime. Use when changing runtime/jobs.py, parse jobs, retries, checkpoints, idempotency keys, leases, conflicts, or recovery behavior.
Implement and review the local Lvke MCP stdio runtime, tool schemas, handlers, Resources, storage, and tests. Use for changes under src/lvkemcp or MCP server behavior. Do not add authentication, roles, tenant isolation, RBAC, permission management, or professional-signoff workflows.
A three-step process for proposing edits, reviewing the differences, and applying the chosen version. It also defines when to stop for user confirmation and how to handle conflicting approval gates.
Implement and verify the Lvke Electron desktop shell and single-machine L3 readiness. Use for desktop packaging, IPC, startup, local runtime, updates, or whole-product desktop verification.
A skill for the Lvke single-machine desktop application and its L3 whole-product readiness checks. It covers Electron application code, startup logic, packaging, Windows installers, code signing, and security fuses.
A standard procedure for handling failures such as locked workspaces, failed language-model gateways, conflicting proposals, or failed financial checks.
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Claude ('No, that's wrong...', 'Actually...'), (3) User requests a capability that doesn't exist, (4) An external API or tool fails, (5) Claude realizes its knowledge…
A financial-modelling guide focused on financing. It covers the timing of funding, ability to repay debt, and whether the capital structure is appropriate, rather than only listing funding sources.
★not rated 0 3d agoA39 tokens
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: