tibursocampos/agent-dev-toolkit / core/skills/developer
developer
UnverifiedSmart router for stack skills, or senior fullstack for ad-hoc scripts/HTML/automation. Use for generic development or when invoking /developer.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.2 KB |
| reference.md | markdown | 533 B |
| references/fallback-execution.md | markdown | 2.5 KB |
| references/frontend-design-routing.md | markdown | 931 B |
| references/must-not.md | markdown | 500 B |
| references/stack-routing.md | markdown | 4.2 KB |
references/fallback-execution.md
View on GitHub## Fallback execution (no stack match)
### Subagent-first (before implement)
Stack `*-developer` skills own this policy when routed. In fallback mode: classify complexity → consult capability `subagents` → if medium/complex and `native`: spawn ≤2 children (scoped **paths** + **receipt**); **trivial** stays **in-parent**; if `subagents=none` or Agent unavailable → **fallback** **in-parent** (never hard-fail). Load `SPAWN.md` + `subagent-first.md`; do not paste guidelines into child prompts.
### 0. Workspace
Confirm target repo, read `README.md` (if exists), and summarize requested acceptance. Before choosing tests or validation, inspect existing test directories, runner/configuration, project scripts, and CI commands; preserve the repository's verified baseline.
For an existing repository, or when impact and local conventions are unclear, follow `{{TOOLKIT_ROOT}}/skills/_shared/agents/BROWNFIELD-ANALYSIS.md` before the first implementation change. Reuse parent-provided analysis only after checking that it is current and covers the target area; re-check changed or uncovered paths. Keep discovery proportional to the affected scope, cite representative local evidence, and surface material gaps rather than choosing toolkit or external examples as defaults. Stack-specific skills retain ownership of their commands and conventions; do not duplicate the full checklist here or in each stack skill.
### 1. Micro-plan
Checkpoint durable workflow progress at the workflow-defined boundaries; context usage does not add a stop gate.
### 2. Implement
Recheck applicable project patterns in the touched area before editing, using the shared brownfield analysis from Step 0. Discover architectural patterns in the touched area first. Follow `structure-and-quality.md` §1: an operator order and the established project style outrank the toolkit style bar. Do not restyle existing members without an explicit yes. On greenfield / no clear comment-language mirror, ask before writing narrative comments/docs (identifiers stay English).
Write clean, maintainable code following universal best practices for the target language (e.g., HTML, Bash, Python script).
### 3. Tests / Validation
Run the project's configured tests, scripts, or linting tools relevant to the change. If no test or validation baseline exists, choose a minimal check appropriate to the task; do not introduce a stack-specific runner by default in generic fallback mode.
### 4. Handoff
Offer `/commit`. Do not commit automatically.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 467k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 319k | Updated | Claude CodeCodex | Verified | Install |