executor · devin-axis — Claude Code subagent: what it does, how to install
executor — a subagent takes a well-defined task from an orchestrator, edits code, runs commands, and reports the result. Use it to delegate implementation work.
LicenseRef-Proprietary ★ 7,523 updated 2026-10-29
.claude/agents/executor.mdWhat executor does
executor handles focused development tasks: it writes or edits code, runs repository commands, and sends back a record of changed files and outcomes. Before reporting, it runs a type check, lint, test, or build when a quick check is available.
The agent follows the repository’s AGENTS.md rules, including pnpm, avoiding any and as casts, keeping the diff small, and preferring @/components with shadcn/ui and Base UI. An underspecified task is a stop condition, not an invitation to guess. Its configuration uses openai/gpt-5.5-fast with the xhigh variant.
What it helps with
- delegate a defined coding task to an orchestrator
- edit project files according to a concrete plan
- run an available build, lint, or test check
- report every touched file and its result
- record assumptions and unfinished follow-ups
Example prompts for the agent
- Ask the orchestrator to send
executora concrete task forsrc/..., then verify the build and report changed files. - Have
executorfix the issue insrc/..., run the available tests, and return the results. - Send
executora narrowly defined repository task and stop if its requirements are ambiguous.
How to install executor
Manually: put the files in place
.claude/agents/executor.mdFor all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention
Manually: put the files in place
.codex/agents/executor.tomlCodex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/
Manually: put the files in place
.cursor/agents/executor.mdCursor also reads .claude/agents/
Step by step, with commands for each platform: Claude Code subagents: how to create one, call it and keep context clean
Questions about executor
How do I install executor in Claude Code?
Create the agent file .claude/agents/executor.md in the project root with the subagent configuration.
What does executor do?
It implements a defined coding task, runs commands and checks, then reports file changes, results, assumptions, and follow-ups to the orchestrator.
What happens when the task is underspecified?
The subagent should flag the ambiguity and stop instead of extending the scope or inventing requirements.
Similar subagents
All subagents for Agent setup: memory, context, orchestration →executor by other authors comes first.