/dev-ultraqa
QA cycling workflow — repeat build/lint/test/fix cycles up to 5 times until all checks pass. Stops early if the same error repeats 3 times (signals a fundamental issue, not a fixable one).
$ npx -y skills add evolution-foundation/evo-nexus --skill dev-ultraqa --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
- Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
- You can call itInvoke it directly when you want it.
- Slash command
/dev-ultraqa
Context preview
The summary Claude sees to decide when to auto-load this skill.
QA cycling workflow — repeat build/lint/test/fix cycles up to 5 times until all checks pass. Stops early if the same error repeats 3 times (signals a fundamental issue, not a fixable one).
SKILL.md
dev-ultraqa.SKILL.mdname: dev-ultraqa
description: QA cycling workflow — repeat build/lint/test/fix cycles up to 5 times until all checks pass. Stops early if the same error repeats 3 times (signals a fundamental issue, not a fixable one).
Dev UltraQA
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
QA cycling: keep running build/lint/test/fix loops until all checks pass — but with a circuit breaker. If the same error persists across 3 cycles, stop and report: it's a fundamental issue, not a fixable one.
Use When
- After a multi-file implementation (post-`dev-autopilot` or `@bolt-executor`)
- Pre-merge gate when you want all checks green before review
- Refactor that touches enough files to risk regressions
Do Not Use When
- Single-file change → just run `dev-verify` once
- The first failure is structural → escalate to `@apex-architect`, don't loop
Workflow
Each cycle:
1. Run **build** (`npm run build`, `cargo build`, `go build`) 2. Run **lint** (`npm run lint`, `cargo clippy`, `golangci-lint`) 3. Run **tests** (`npm test`, `cargo test`, `pytest`) 4. **If green:** done, exit cycle, hand off to `@oath-verifier` for final verification 5. **If red:** delegate to `@bolt-executor` or `@hawk-debugger` to fix the failure 6. **Loop** back to step 1
Circuit breaker (3-failure rule)
If the same error fingerprint (file:line + error message) appears in 3 consecutive cycles:
- STOP the loop
- Hand off to `@apex-architect` with full context dump
- The issue is no longer "fix the bug" — it's "question the architecture"
Max iterations
- Default: 5 cycles
- After 5 cycles even with different errors: stop and report
Output
Saved to `workspace/development/verifications/[C]ultraqa-{component}-{date}.md`:
## UltraQA Report
### Cycles
1. Build ❌ → fixed by @bolt-executor
2. Build ✅ Lint ❌ → fixed by @bolt-executor
3. Build ✅ Lint ✅ Tests ❌ → fixed by @hawk-debugger
4. ✅ All green
### Final State
- Build: ✅
- Lint: ✅
- Tests: ✅ N passed, 0 failed
### Handoff
- → @oath-verifier for evidence-based verification
Pairs With
- `@bolt-executor` (fixes during cycles)
- `@hawk-debugger` (fixes when bugs are non-trivial)
- `@oath-verifier` (final verification post-cycle)
- `@apex-architect` (escalation when 3-failure breaker trips)
Read more
name: dev-ultraqa description: QA cycling workflow — repeat build/lint/test/fix cycles up to 5 times until all checks pass. Stops early if the same error repeats 3 times (signals a fundamental issue, not a fixable one).
Dev UltraQA
Derived from oh-my-claudecode (MIT, Yeachan Heo). Adapted for the EvoNexus Engineering Layer.
QA cycling: keep running build/lint/test/fix loops until all checks pass — but with a circuit breaker. If the same error persists across 3 cycles, stop and report: it's a fundamental issue, not a fixable one.
Use When
- After a multi-file implementation (post-`dev-autopilot` or `@bolt-executor`)
- Pre-merge gate when you want all checks green before review
- Refactor that touches enough files to risk regressions
Do Not Use When
- Single-file change → just run `dev-verify` once
- The first failure is structural → escalate to `@apex-architect`, don't loop
Workflow
Each cycle:
1. Run **build** (`npm run build`, `cargo build`, `go build`) 2. Run **lint** (`npm run lint`, `cargo clippy`, `golangci-lint`) 3. Run **tests** (`npm test`, `cargo test`, `pytest`) 4. **If green:** done, exit cycle, hand off to `@oath-verifier` for final verification 5. **If red:** delegate to `@bolt-executor` or `@hawk-debugger` to fix the failure 6. **Loop** back to step 1
Circuit breaker (3-failure rule)
If the same error fingerprint (file:line + error message) appears in 3 consecutive cycles:
- STOP the loop
- Hand off to `@apex-architect` with full context dump
- The issue is no longer "fix the bug" — it's "question the architecture"
Max iterations
- Default: 5 cycles
- After 5 cycles even with different errors: stop and report
Output
Saved to `workspace/development/verifications/[C]ultraqa-{component}-{date}.md`:
## UltraQA Report ### Cycles 1. Build ❌ → fixed by @bolt-executor 2. Build ✅ Lint ❌ → fixed by @bolt-executor 3. Build ✅ Lint ✅ Tests ❌ → fixed by @hawk-debugger 4. ✅ All green ### Final State - Build: ✅ - Lint: ✅ - Tests: ✅ N passed, 0 failed ### Handoff - → @oath-verifier for evidence-based verification
Pairs With
- `@bolt-executor` (fixes during cycles)
- `@hawk-debugger` (fixes when bugs are non-trivial)
- `@oath-verifier` (final verification post-cycle)
- `@apex-architect` (escalation when 3-failure breaker trips)
Other skills on evo-nexus.
- /ai-image-creator
Generate PNG images using AI (multiple models via OpenRouter including Gemini, FLUX.2, Riverflow, SeedDream, GPT-5 Image, proxied through Cloudflare AI Gateway BYOK). Also analyze/describe existing images using multimodal AI vision. Use when user asks to "generate an image",
Open skill - /create-agent
Create a new custom agent for the workspace. Guides the user through defining agent name, domain, personality, skills, model, and memory folder. Use when the user says 'create an agent', 'new agent', 'add an agent', 'I need a custom agent', or wants to create a specialized agent
Open skill - /create-command
Create a new slash command for Claude Code. Guides the user through defining the command name, what it does, and generates the markdown file in .claude/commands/. Use when the user says 'create a command', 'new command', 'add a slash command', 'I want a shortcut for', or wants
Open skill - /create-goal
Create a Mission, Project, or Goal (Mission → Project → Goal → Task hierarchy) in EvoNexus. Guides the user through picking a mission, choosing or creating a project, defining a measurable goal with metric_type and target_value. Writes to the SQLite goals tables via POST
Open skill - /create-heartbeat
Create a new heartbeat (proactive agent scheduled with a decision prompt) for EvoNexus. Guides the user through picking an agent, setting interval, wake triggers, and the decision prompt that governs when the agent acts. Writes to config/heartbeats.yaml with pydantic validation.
Open skill - /create-integration
Create a new custom integration (API/service wrapper) for the workspace. Guides the user through defining the integration's slug, display name, description, category, and required env keys. Writes .claude/skills/custom-int-{slug}/SKILL.md via POST /api/integrations/custom. Use
Open skill

