vinilana

Implement or review bounded MCP inputs and responses. Use for gateway response helpers, MCP Zod schemas, list actions, cursor pagination, maxEvents, JSON serialization, audit logging, response metadata, artifact exports, or payload-size and…

Implement or review child-process execution with bounded stdout and stderr memory. Use for Node spawn wrappers, sensors, test runners, acceptance commands, timeout handling, output tails, truncation, or any change that captures subprocess output in the…

Design, implement, or review bounded cache and persistent-index lifecycle. Use for ContextCache, SemanticContextBuilder, tree-sitter analysis cache, MCP session cache, hook host-session bindings, checkpoints, TTL, LRU, byte budgets, invalidation, cleanup…

Implement or review language-server child-process ownership and cleanup. Use for LSPLayer, optional semantic LSP mode, spawn initialization, request timeouts, buffers, pending requests, retries, circuit breakers, graceful shutdown, SIGTERM/SIGKILL escalation,…

Design, implement, or review safe lifecycle-hook payload capture and trace persistence. Use for Claude Code, Codex, or Pi hook dispatch; PostToolUse Write/Edit/Bash mapping; stdin parsing; trace redaction; trace quotas; rotation; or hook privacy and…

Consolidate or review repository discovery, semantic parsing, functional-pattern detection, codebase fingerprints, and snapshot generation. Use for context init, DocumentationGenerator, AgentGenerator, SemanticSnapshotService, CodebaseAnalyzer, FileMapper,…

Implement or review bounded runtime-history storage and retrieval. Use for traces, sessions, artifacts, sensor summaries, replays, datasets, cursor pagination, JSONL streaming, pruning, retention, or any harness API whose cost can grow with historical records.

Review code quality, patterns, and best practices

Add and enforce Open Fusion operational guardrails. Use when Codex works on request ids, structured logging, metrics, health checks, authentication, authorization, payload limits, timeout/retry policy, redaction, provider failure handling, rate limiting, or…

Build and maintain Open Fusion provider adapters around the Vercel AI SDK. Use when Codex adds or changes OpenRouter integration, provider adapter interfaces, provider-specific configuration, model capability mapping, Vercel AI SDK calls, provider error…

Verify Open Fusion changes before completion. Use when Codex is finishing a task, preparing a PR, reviewing readiness, or checking software quality gates such as tests, lint, typecheck, formatting, docs alignment, security redaction, SOLID/DRY risks, and…

Design the right testing approach for Open Fusion. Use when Codex needs to choose, write, restructure, or review unit, integration, contract, e2e, streaming, provider-adapter, config, orchestration, or security tests for the Open Fusion LLM gateway.

Maintain Open Fusion product and architecture documentation. Use when Codex needs to create, update, review, or reconcile the PRD, specs, ADRs, architecture decisions, roadmap scope, API contracts, configuration requirements, provider strategy, or development…

Keep Open Fusion code maintainable with SOLID, DRY, clean architecture, and pragmatic TypeScript/NestJS design. Use when Codex designs modules, writes production code, refactors, reviews code quality, extracts abstractions, names services/adapters, or reduces…

Implement and maintain Open Fusion's single JSON configuration system. Use when Codex works on config schema, loading, validation, environment variable secret resolution, provider/model/route/auth configuration, default config examples, config tests, or…

Implement and maintain Open Fusion's LLM orchestrator workflow. Use when Codex works on orchestrator prompts, route selection, delegate_llm tool behavior, allowed delegate model enforcement, delegation limits, orchestration loops, final response synthesis, or…

Use when designing APIs, tool schemas, commands or integration contracts.

Use when diagnosing errors, regressions or unexpected behavior.

Use when auditing code quality, correctness and maintainability.

Use when preparing concise conventional commit messages and release notes.

Use when creating, updating or reviewing project documentation.

Use when decomposing feature requests into implementation phases.

Use when reviewing a pull request, diff or merge readiness.

Use when restructuring code safely while preserving behavior.

Design RESTful APIs following best practices

Systematic bug investigation and root cause analysis

Review code quality, patterns, and best practices

Generate commit messages following conventional commits with scope detection

Generate and update technical documentation

Break down features into implementable tasks

Review pull requests against team standards and best practices

Safe code refactoring with step-by-step approach

Prepare Invokta versioned releases on dedicated release branches with synchronized package metadata, changelog and validation evidence, complete repository gates, and one cohesive commit. Use when asked to prepare, cut, stage, or validate an Invokta release…

Add or change domain capabilities in this Invokta Action Engine with explicit contracts, authenticated access, engine-owned dependencies, migrations, and tests. Use when a user asks to add an action, tool, capability, domain behavior, repository operation,…

Deploy, upgrade, verify, or roll back this Docker Compose MCP OAuth stack on a VPS, including Hostinger environments with an existing Traefik proxy and hosts that use bundled Caddy. Use for production releases, new VPS setup, image rebuilds, environment…

Develop this generated Invokta Action Engine when changing capabilities, dependencies, tests, or its direct invocation, CLI, MCP stdio, and MCP HTTP channels. Use for implementation, refactoring, debugging, and contract review in this project.

Diagnose, test, and change the self-hosted OAuth layer for this MCP service while preserving PKCE, DCR, ES256 tokens, rotating refresh tokens, secure consent, sanitized logs, and fail-closed resource authentication. Use for OAuth login or consent failures,…

Review Invokta contracts for public APIs, acceptance criteria, errors, invariants, operational limits, compatibility, and architectural boundaries. Use before implementing or approving changes to interfaces, schemas, ports, events, observable persistence,…

Deliver Invokta milestones and work items with TDD, adherence to versioned contracts and architecture, complete validation, and one cohesive commit per work item. Use when developing features, fixing bugs, executing roadmap milestones, or completing…

Always use when planning, designing, scoping, estimating, or proposing any Bench My Harness feature, refactor, adapter, schema, CLI command, benchmark behavior, test suite, ADR, or implementation approach. Enforces Spec Driven Development, TDD-first planning,…

Always use when reviewing, validating, auditing, or summarizing completed Bench My Harness work, including code changes, tests, docs, ADRs, schemas, adapters, CLI behavior, benchmark runner behavior, or pull-request style review. Enforces TDD acceptance…

Write, review, or update Bench My Harness architecture decision records for hexagonal architecture, event schemas, benchmark protocol, adapter strategy, storage, security, comparability, metrics, or execution isolation.

Create, review, or update Bench My Harness benchmark scenarios, suites, prompts, fixtures, expected outputs, validation commands, scoring rubrics, reproducibility metadata, and benchmark YAML or JSON contracts.

Design, implement, or review Bench My Harness adapters for agentic coding harness hooks, plugins, extensions, transcripts, JSONL logs, stdin events, or webhooks from Codex, Claude Code, Cursor, OpenCode, Pi, or custom harnesses.

Analyze Bench My Harness normalized events, raw hook references, metric observations, token usage, context window usage, tool calls, command execution, outputs, benchmark run artifacts, and harness comparison reports.