| name | dev-coder |
| description | This skill should be used when the user or an agent asks to "implement a task", "execute a dev task", "code the task", "write the code for a task", "pick up a task from the orchestrator", or when a task file needs to be implemented. Reads a structured task specification file written by an orchestrator agent and produces working code in the target programming language. |
| version | 1.0.0 |
Dev Coder Skill
Implements coding tasks assigned by an orchestrator agent. Reads a structured task file, produces working code in the specified language, and writes output to the target path.
Supported Languages
Java, Python, JavaScript, TypeScript, C#, Go, Rust, Kotlin, Ruby, Bash/Shell
Task File Location
Look for the task file in this priority order:
- Path given in
$ARGUMENTS (if provided)
.claude/tasks/current-task.json
.claude/tasks/current-task.md
Task File Format (JSON)
{
"task_id": "task-001",
"title": "Short task title",
"description": "What to build and why",
"language": "python",
"output_path": "teams/team-name/exercise_1/filename.py",
"requirements": [
"Specific behaviour or constraint 1",
"Specific behaviour or constraint 2"
],
"context": "Optional background — neighbouring code, data model, etc.",
"assigned_by": "orchestrator-agent",
"dependencies": []
}
Task File Format (Markdown)
# Task: <title>
**ID:** task-001
**Language:** python
**Output:** teams/team-name/exercise_1/filename.py
**Assigned by:** orchestrator-agent
## Description
What to build and why.
## Requirements
- Requirement one
- Requirement two
## Context
Optional background information.
Execution Steps
Step 1 — Read the task
- Locate and read the task file (see priority order above).
- If no task file is found, report the missing file path and stop.
- Confirm understanding: restate the title, language, output path, and requirements in one short paragraph.
Step 2 — Explore context
- If
output_path already exists, read it to understand existing patterns.
- If
context references other files, read those too.
- If the task has
dependencies listing other output files, read each one.
Step 3 — Implement
- Write clean, working code that satisfies every listed requirement.
- Match the conventions of any existing code found in Step 2.
- Do NOT add features beyond what the requirements specify.
- Do NOT add comments unless the logic is genuinely non-obvious.
Language-specific defaults:
- Java — JDK 17+, standard Maven layout, JUnit 5 for tests
- Python — 3.10+, type hints, no third-party deps unless stated
- TypeScript — strict mode, ESM imports
- JavaScript — ESM, no transpiler unless stated
- C# — .NET 8, nullable enabled
- Go — modules, idiomatic error returns
- Rust — 2021 edition, Result/Option idioms
- Kotlin — JVM target, coroutines if async needed
- Ruby — 3.x, frozen string literal
- Bash —
set -euo pipefail, POSIX-compatible unless stated
Step 4 — Write output
- Write the code to
output_path specified in the task file.
- If the output path does not exist yet, create any needed parent directories first.
Step 5 — Report completion
Write a brief completion summary containing:
task_id and title
output_path of the file written
- A bullet list of each requirement and whether it was satisfied
- Any assumptions made that were not explicit in the task
Error Handling
| Problem | Action |
|---|
| Task file not found | Report the expected path; do not guess or create a placeholder task |
| Language not supported | Report clearly; ask the user to add it to the skill |
output_path directory missing | Create it, then write the file |
| Conflicting requirements | List the conflict and ask for clarification before writing code |
| Ambiguous requirement | State your interpretation, implement it, and flag it in the completion report |
Notes
- Never modify the task file itself — it is owned by the orchestrator agent.
- Never invent requirements not present in the task file.
- If the task file contains a
dependencies list, all dependency files should exist before implementation. If they do not, report the missing dependency and stop.