Skip to content
Agentmart
Submit a tool

dev

Development entry point — routes a request to the right workflow.

Skill by charleshall888

IndexedMIT★ 3
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/charleshall888-dev

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Development entry point — routes a request to the right workflow. Use for "what should I work on", "what's next", or a feature described without naming a skill.

Use cases: productivity

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 26, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/dev

What's inside

Skill name
dev
Description
Development entry point — routes a request to the right workflow. Use for "what should I work on", "what's next", or a feature described without naming a skill.
License
MIT
Instructions
49 lines in SKILL.md · view on GitHub

Files (2)

  • SKILL.md
  • size-pin.txt · references/

SKILL.md (first lines)

# Dev

Route a development request to the right workflow. A skill named in the request is a strong hint, not an order: judge the route yourself and say if you disagree.

## Step 1: Route

First match wins.

**Before matching on an explicit ticket id**, run `cortex-lifecycle-next <id>`. `closed` or `parked` means an outcome is already recorded. Show it with its reason and ask whether to reopen or unpark. Route only on a yes.

1. **No arguments, or "what should I work on" / "what's next"** → backlog triage (Step 3).
2. **Three or more features, or a batch** → classify each by rule 4. All above simple → `/cortex-overnight:overnight` with the list. All simple → implement each here. Mixed → show a task/routing/justification table and confirm.
3. **Vague topic** ("not sure how to approach", "explore", "investigate") → `/cortex-core:discovery <topic>`.
4. **`simple`** — you know the approach, or one read confirms it, and nothing needs deciding. Size is not the test. → Implement here and commit. Close any backlog item it resolved with `cortex-update-item {slug} --status complete` (skip with no item or an external backend). Do not write `--complexity`: nothing reads a tier recorded outside Clarify.
5. **Otherwise** → assess criticality (Step 2), then route on readiness:
   - no `spec:` (or no ticket) → `/cortex-core:refine <feature-name>` with the criticality context
   - `status: refined` with a `spec:` → `/cortex-core:build <feature-name>`
   - unsure → `cortex-lifecycle-next <feature>` reports the served phase: `research`/`specify` means refine, later means build

Plan mode records nothing in the lifecycle. A plan for routed work is build's plan phase.

## Step 2: Criticality

If `cortex/lifecycle/<feature>/` exists, read `cortex-lifecycle-state --feature <feature> --field criticality` and ask whether to resume at its served phase. A fresh start needs the user to confirm that existing artifacts are discarded.

Otherwise suggest:

- **critical** — security, money, or data loss
- **high** — shared libraries, CI/CD, migrations, foundational tooling, or a broad change that is hard to reverse
- **low** — docs and formatting
- **medium** — anything else, or when uncertain

Present it as **Criticality suggestion: `<level>`** — `<one-sentence justification>`.

## Step 3: Backlog triage

```bash
cortex-backlog-triage
```

This one call also rebuilds the index. By `state`:

- `ok` → print `blocks` exactly as given. Ask which item or parallel-safe group to pick up, then route it from Step 1. For an `idea` row, follow the printed recommendation. A multi-item pick goes through rule 2.
- `external-backend` → show `message`. The local index is not the source of truth, so route through refine or discovery and leave it alone.
- `no-index` / `error` → show `message` and stop.

If the user overrides a suggested route, their choice wins at once. If the scope changes, classify again from Step 1.

From charleshall888/cortex-command, MIT.

More from charleshall888/cortex-command

Signals

Stars
3
Forks
0
Open issues
0
Last commit (this folder)
Sep 22, 2026 (1 month ago)
Last release
Oct 5, 2026 (24 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from charleshall888

Cite this page

dev skill. Agent Mart. https://agentmart.dev/skills/charleshall888-dev (last updated 2026-10-24)