Use when the user returns to a session and needs to be reoriented, including requests such as "amnesia", "what did I miss", "recap the session", "where were we", "catch me up", "what happened while I was away", or "what was I doing". Reports when the session started, the last user message verbatim, what happened since…
Use when the codex CLI should be driven from the agent — as a reviewer giving a second, independent opinion ("ask codex", "call codex", "codex review", "get a second opinion", "challenge this finding"), or as an executor doing a scoped task in the repository ("have codex do this", "delegate this to codex", "let codex…
Audit, prune, rewrite, or relocate Claude Code project memory. Use when the user asks to review, clean, update, or remove Claude memory files, eliminate stale project memories, or rebuild MEMORY.md.
Audit and reduce low-value code comments, KDoc, docstrings, CLAUDE.md, and AGENTS.md while preserving behavior and durable intent. Use for repository-wide comment cleanup, comment-density reduction, or shortening agent guidance.
Use when fixing software issues in a repository, including bug reports, failing tests, CI failures, compile errors, runtime errors, regressions, dependency problems, and requests such as "fix this issue", "resolve this bug", "make tests pass", or "address the failure". Guides Codex through issue triage, reproduction…
Send a message from an external agent or script to a running Claude Code session on the same machine. Use when a manually started Codex session needs to tell any Claude session something, Codex needs to report back to the Claude session that launched it, a hook or long job should notify Claude, or the user asks to…
Queue a message from Claude Code, another agent, or a script to an existing Codex session. Use when Claude needs to report back to Codex or the user asks to "send this to Codex", "tell another Codex session", "notify the Codex session", or "queue a message for Codex". Not for starting a separate Codex review or task…
Use when building, modifying, debugging, or reviewing Ktor 3.5.x server and client applications, including routing, requests/responses, plugins, serialization, authentication, sessions, WebSockets, SSE, engines, testing, deployment, and Kotlin multiplatform client work.
Use when creating, writing, or running standalone executable Kotlin scripts with the .main.kts extension, including the kotlin script runner, shebang execution, @file:DependsOn/@file:Repository/@file:Import/@file:CompilerOptions/@file:OptIn annotations, script dependencies, command-line args, compiled-script caching…
Use when writing, reviewing, or modernizing Kotlin code with recent language versions (2.0–2.4.x, including 2.4.10 and 2.4.20-Beta1), including context parameters, collection literals, guard conditions in when, multi-dollar interpolation, non-local break/continue, explicit backing fields, nested type aliases…
Create focused kotgent backlog tasks from a user request, specification, issue list, or implementation plan, with repository and session provenance. Use when the user asks to create, add, file, split, or seed tasks in kotgent; do not use for comparing or reviewing an existing backlog.
Audit todo kotgent tasks against the current repository and recommend whether to keep, rewrite, merge, drop, or escalate them. Use when the user asks to review, groom, clean up, triage, or validate the todo backlog for completed, obsolete, conflicting, duplicate, absorbed, or incorrectly blocked tasks. This workflow…
Compare active kotgent tasks with repository guidance, plans, code, tests, TODOs, and Git history, and report task/repository drift with evidence and proposed CLI commands. Use when the user asks to sync, reconcile, map, or compare the kotgent backlog with the current repository. This workflow is always read-only.
Safely claim a specific or next eligible kotgent task from a free session, detect blocked or concurrent work, and immediately continue through the work-task workflow. Use when the user asks to take, claim, pick up, start, or work the next kotgent task, optionally naming a task ref.
Execute the kotgent task linked to the current session, coordinate concurrent work, implement and verify repository changes, report meaningful progress or blockers, and submit the task for human review. Use when the user asks to work, continue, implement, finish, or resume a linked kotgent task.