korchasa

Use when the user asks for a project health audit, maintenance sweep, or multi-category lead-engineer scan followed by interactive issue-by-issue resolution with user approval. Do NOT trigger on routine lint/test runs, single-file cleanups, or standard "check project" requests.

Create, maintain, and run evidence-based benchmarks for AI agents. Use when setting up testing infrastructure, writing new test scenarios, or evaluating agent performance.

Generate a .devcontainer (devcontainer.json plus optional Dockerfile) for AI-IDE development with AI CLI integration, skill mounting, and security hardening. Use when the user wants to set up or containerize a dev environment for Claude Code, OpenCode, or flowai.

Guide for creating effective Agent Skills (SKILL.md packages). Use when users want to create a new skill, write a skill, author a SKILL.md, or ask about skill structure, best practices, or skill file format. Works across IDEs (Cursor, Claude Code, OpenCode).

Estimate total token usage and cost across conversation history, system prompts, and active rules. Use when the user asks how much context or how many tokens are in use, to audit context size, or to estimate session cost.

Design and build plugins for AI IDEs (skills, MCP tools, hooks, packaging, manifests, marketplaces, per-IDE validation). Use when the user wants to create, design, or package an AI-IDE plugin.

Build AI-IDE plugin hook elements - shared policies plus Claude Code/Codex/OpenCode/Cursor adapters for lifecycle events, blocking, audit, and context injection. Use when adding or configuring plugin hooks.

Design AI-IDE plugin marketplaces - requirements, host-specific payloads, local dogfood, release sync, and validation. Use when the user wants to create or publish a plugin marketplace.

Create, maintain, and run evidence-based benchmarks for AI agents. Use when setting up testing infrastructure, writing new test scenarios, or evaluating agent performance.

Use when the user asks for a project health audit, maintenance sweep, or multi-category lead-engineer scan followed by interactive issue-by-issue resolution with user approval. Do NOT trigger on routine lint/test runs, single-file cleanups, or standard "check project" requests.

Guide for creating effective Agent Skills (SKILL.md packages). Use when users want to create a new skill, write a skill, author a SKILL.md, or ask about skill structure, best practices, or skill file format. Works across IDEs (Cursor, Claude Code, OpenCode).

Estimate total token usage and cost across conversation history, system prompts, and active rules. Use when the user asks how much context or how many tokens are in use, to audit context size, or to estimate session cost.

Design and build plugins for AI IDEs (skills, MCP tools, hooks, packaging, manifests, marketplaces, per-IDE validation). Use when the user wants to create, design, or package an AI-IDE plugin.

Build AI-IDE plugin hook elements - shared policies plus Claude Code/Codex/OpenCode/Cursor adapters for lifecycle events, blocking, audit, and context injection. Use when adding or configuring plugin hooks.

Design AI-IDE plugin marketplaces - requirements, host-specific payloads, local dogfood, release sync, and validation. Use when the user wants to create or publish a plugin marketplace.

Build AI-IDE plugin MCP elements - stdio JSON-RPC servers, tool schemas, host wiring for Claude Code/Codex, event-name mapping, and validation. Use when adding an MCP server to a plugin.

Analyze and answer user questions in Autonomous Mode by reading documentation and inspecting codebase

Automated commit workflow with atomic grouping

Initialize project with AGENTS.md and rules, handling both Greenfield (new) and Brownfield (existing) projects.

Iterative issue investigation with user-controlled hypothesis selection

Perform a comprehensive "Lead Engineer" audit: structure, consistency, code quality, technical debt, documentation coverage, and terminology checks.

Create critiqued plan in whiteboard.md using GODS framework with proactive uncertainty resolution

Analyze agent's process, logic, technical decisions, context usage, and undocumented discoveries to find behavioral errors, poor engineering choices, inefficiencies, and missing knowledge in project instructions.

Composite command: review changes then commit only if approved. Inlines flow-review and flow-commit with a verdict gate between them.

Launch FoxCode in Project Profile mode from Codex via the foxcode MCP server (`launchBrowser`). One tool call, no Python.

Launch FoxCode in Project Profile mode via the foxcode MCP server (`launchBrowser` tool). One tool call, no Python.

Test FoxCode distribution when asked to verify install/update paths, plugin packaging, marketplace contents, npm pack output, Claude Code plugin delivery, Codex plugin/config delivery, OpenCode package delivery, release readiness, or user-machine installation behavior.

Launch FoxCode in User Profile mode from Codex. Guides extension loading via about:debugging, opens connection page, verifies connectivity.

Quality-test the user-facing FoxCode skills (`foxcode-run-project-profile`, `foxcode-run-user-profile`) by spawning fresh subagents per probe so context cannot leak. Judges triggering accuracy, end-state correctness, robustness on negative prompts, and adherence to the SKILL.md flow. Use when asked to QA, audit, benchmark, regression-check, or smoke-test FoxCode skills, or to validate that a SKILL.md edit did not break the run flow.

Launch FoxCode in User Profile mode. Guides user to load extension via about:debugging, opens connection page, verifies connectivity.

Analyze historical Codex and Claude Code sessions for real FoxCode usage. Use when evaluating how effectively agents use FoxCode, evalInBrowser, launch skills, or browser automation across past sessions.

Run FoxCode acceptance testing when asked to verify that FoxCode works end-to-end, smoke-test browser automation, validate evalInBrowser, check Claude Code/OpenCode/Codex integration, or prove a release candidate works with real Firefox.

Supervisor for one flowai-workflow run (Codex). Starts or resumes a single workflow, diagnoses failures from run artifacts, patches root causes, resumes the same run. Run inside an isolated Codex worker subagent spawned by the `supervise`/`orchestrate` skill.

Copy a bundled flowai-workflow into the user's project so they can adapt and run it locally. Use when the user wants to start using flowai-workflow in a new repo.

Run the flowai-workflow orchestration policy loop. Delegates policy execution to the workflow orchestrator agent, which selects workflows from ORCHESTRATION.md and supervisor summaries.

Execute a bundled or project-local flowai-workflow DAG. Use to launch a workflow run from inside the host IDE.

Live flowai-workflow run supervisor. Use only for requests to run, monitor, continue, or resume a current workflow now. Exclude last-night, finished, post-mortem, root-cause-report, and conceptual questions.

Long-cycle flowai-workflow orchestrator (Codex). Reads project orchestration policy, selects the next workflow, and returns a structured supervisor delegation request. Run inside an isolated Codex worker subagent spawned by the `orchestrate` skill.

Scaffold or adapt flowai-workflow DAGs in a project. Use for adding flowai-workflow, adapting an existing .flowai-workflow/<name>, validating workflow.yaml, or wiring agent prompts/scripts.

Set up flowai-workflow engine in a project: create workflow YAML config, agent definitions, directory structure, and run scripts. Use when the user wants to add flowai-workflow to a project, create a new workflow, configure DAG-based agent pipelines, or asks about workflow setup/configuration.

Create GitHub issue as a requirement ticket: problem + requirements only, no implementation details. Use when user asks to create a ticket, issue, or task for a feature/fix/change.

Use when the user asks for a project health audit, maintenance sweep, or multi-category lead-engineer scan followed by interactive issue-by-issue resolution with user approval. Do NOT trigger on routine lint/test runs, single-file cleanups, or standard "check project" requests.

Collects Claude Code, OpenCode, Cursor, and Codex release changes for a requested time window and evaluates relevance to @korchasa/ai-ide-cli. Use when asked to monitor IDE/agent CLI updates, compare versions, or assess upgrade impact.

Generates bg-trainer game data (DataItem arrays, mode definitions, lesson registration) from a Bulgarian textbook lesson file at documents/lessons/lesson-N.md. Use when the user asks to cover / port / add / generate games (modes, drills, exercises) for a specific lesson file, enable a lesson marked "Скоро", or extend an existing lesson's mode coverage. Encodes project-specific engine selection, item schemas, i18n (ru/uk) rules, and gotchas learned from post-release fixes (disambiguation, Lq "/" split, no-context judgments, LiEngine focus markers, rule/hint/label separation).