engineering-standards-orchestrator
Coordinate programming work through mandatory SOLID architecture, test-driven development, Loop Engineering, and completion-quality gates.
Skill by projectamazonph
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/projectamazonph-engineering-standards-orchestrator
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.
About
Use cases: design, testing, devops, productivity
Reviews
Write a reviewNo 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. Checked Oct 25, 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/engineering-standards-orchestrator
What's inside
- Skill name
engineering-standards-orchestrator- Description
- Coordinate programming work through mandatory SOLID architecture, test-driven development, Loop Engineering, and completion-quality gates. Use for any task that designs, creates, modifies, fixes, refactors, migrates, reviews, or verifies source code, tests, schemas, configuration, build logic, CI workflows, APIs, or software architecture. Also use for coding plans that will guide later implementation. Route focused work to the installed solid-code-design, tdd-workflow, loop-engineering, and engineering-quality-gates skills at the required moments.
- License
- MIT
- Instructions
- 168 lines in SKILL.md · view on GitHub
Files (2)
- SKILL.md
- pack-contract.md · references/
SKILL.md (first lines)
# Engineering Standards Orchestrator ## Mission Keep engineering standards active from intake through completion. Own sequencing, routing, state, and enforcement. Do not duplicate the specialists' detailed methods. Treat “done” as a verified state, not a confident sentence. ## Coordinate the Pack Apply each specialist only at its responsibility boundary: | Skill | Invoke at | Required return | | --- | --- | --- | | `loop-engineering` | Start of every non-trivial implementation, fix, refactor, or migration; keep active throughout | Current loop, evidence, outcome, and next action | | `solid-code-design` | Before adding or changing architectural boundaries; during design review; when test friction exposes coupling | Change map, boundary decision, contracts, dependency direction, and architecture risks | | `tdd-workflow` | Before changing production behavior or fixing a defect | Valid Red evidence, minimal Green change, and focused test result | | `engineering-quality-gates` | After the requested behavior is complete and before reporting completion | PASS or BLOCKED with evidence and remediation route | Read [references/pack-contract.md](references/pack-contract.md) when a task spans several modes, a specialist handoff is ambiguous, or a gate fails. ## Maintain the Control Card Maintain this compact state internally or in the repository's existing engineering log. Surface it only when useful to the user. ```text Mode: design | build | fix | refactor | migrate | review Risk: low | medium | high Phase: intake | baseline | shape | red | green | refactor | observe | gate | complete | blocked Active loop: <one behavior or structural move> Architecture decision: <boundary or none needed> Red evidence: <command and expected failure> Green evidence: <command and pass result> Observation: <broader checks and runtime evidence> Gate: pending | pass | blocked Next action: <smallest justified step> ``` Update the card at every phase transition. Never carry a guessed or stale status forward. ## Run the State Machine ### 1. Intake 1. Read repository instructions and user constraints. 2. Classify the mode and risk. 3. Inspect the working tree before planning edits; preserve unrelated changes. 4. Define the requested outcome, observable acceptance behavior, and out-of-scope work. 5. Start `loop-engineering` for non-trivial work. ### 2. Establish the Baseline 1. Inspect relevant code, callers, tests, interfaces, and dependency edges. 2. Run the smallest relevant existing checks. 3. Record pre-existing failures separately from failures introduced by the task. 4. Stop and report when the baseline cannot be established and proceeding would risk destructive or misleading work. ### 3. Shape the Change Invoke `solid-code-design` when any of these conditions apply: - Add a class, module, service, interface, adapter, repository, strategy, plugin, or inheritance relationship. - Move responsibilities or invert a dependency. - Change a public contract, schema, API boundary, or cross-module data flow. - Introduce another implementation or variation point. - Encounter a test that requires excessive setup, irrelevant mocks, real infrastructure, or subtype exceptions. Skip a new abstraction when the change is local, cohesive, and has no credible variation or infrastructure boundary. Record “no new boundary needed” as a valid architecture decision. ### 4. Drive Behavior Test-First Invoke `tdd-workflow` before the first production-code change for every behavior addition or defect repair. Require this order: 1. Select one observable behavior. 2. Prove a valid Red state for the intended reason. 3. Make the smallest Green change. 4. Refactor only while tests remain green. Do not accept a syntax error, broken fixture, unrelated failure, skipped test, or weakened assertion as Red or Green evidence. For a behavior-preserving refactor, require characterization or contract tests before moving structure. For a pure review, do not mutate files and do not manufacture a TDD cycle. ### 5. Continue Tight Loops Return control to `loop-engineering` after each Green state or structural move. Require the loop to: 1. Observe focused and proportionate broader evidence. 2. Compare the result with the hypothesis and acceptance behavior. 3. Record the outcome. 4. Choose the next smallest behavior, structural move, or diagnostic action. Split any loop that contains several independently verifiable outcomes. Diagnose unexpected failures before adding more changes. ### 6. Enforce Completion Invoke `engineering-quality-gates` only after all requested acceptance behavior appears complete. Do not report completion unless the gate returns PASS. Route BLOCKED findings back to the appropriate phase: | Finding | Return to | | --- | --- | | Missing or invalid behavior proof | `tdd-workflow` at Red | | Coupled or unjustified design | `solid-code-design` at Shape | | Regression, build, lint, type, or runtime failure | `loop-engineering` at Observe/Adjust | | Scope drift or unrelated diff | Intake/Shape | | Missing required migration, security, or rollback evidence | Baseline/Shape | Re-run the gate after remediation. Never self-approve an unresolved required finding. ## Adapt by Task Mode ### Design or Planning Only Use `solid-code-design` to define change axes, contracts, dependency direction, and test seams. Use the quality gate to check that the plan contains acceptance criteria and verification strategy. Do not pretend tests were run.
From projectamazonph/agent-skill-framework, MIT.
More from projectamazonph/agent-skill-framework
- 0.529score
loop-engineering-v2
projectamazonph
Execute non-trivial software work through small, observable build-test-observe-adjust loops with explicit hypotheses, evidence, outcomes, and next actions.
SkillBFreshness: BClaude Code +9 - 0.529score
engineering-quality-gates
projectamazonph
Verify engineering work against required behavioral, architectural, regression, build, security, compatibility, documentation, and diff-hygiene evidence…
SkillBFreshness: BClaude Code +9 - 0.529score
solid-code-design
projectamazonph
Design, refactor, and review object-oriented or modular code with the SOLID principles while preserving behavior and avoiding needless abstraction.
SkillBFreshness: BClaude Code +9 - 0.529score
tdd-workflow
projectamazonph
Drive one observable software behavior at a time through a valid Test-Driven Development cycle: select the test, prove Red for the intended reason, implement…
SkillBFreshness: BClaude Code +9 - 0.529score
solid
projectamazonph
Use this skill when writing code, implementing features, refactoring, planning architecture, designing systems, reviewing code, or debugging.
SkillBFreshness: BClaude Code +9 - 0.523score
precision-workflow
projectamazonph
execute complex work with research, files, and quality gates
SkillBFreshness: BClaude Code +9 - 0.521score
amazon-ad-console-redesign
projectamazonph
Complete premium redesign of the Amazon Ad Console Training Simulator using the Design Skill Pack.
SkillBFreshness: BClaude Code +9 - 0.521score
design-system-master
projectamazonph
Master design system orchestrator — selects, combines, and applies the right design skills from the pack based on the project context.
SkillBFreshness: BClaude Code +9 - 0.507score
project-bootstrap-full
projectamazonph
Create a complete project foundation with SOLID architecture, TDD protocols, comprehensive documentation suite, data logic wireframes, and project scaffold.
SkillBFreshness: BClaude Code +9 - 0.499score
design-an-interface
projectamazonph
Generate multiple radically different interface designs for a module using parallel sub-agents.
SkillBFreshness: BClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Aug 13, 2026 (2 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from projectamazonph
- 0.529score
loop-engineering-v2
projectamazonph
Execute non-trivial software work through small, observable build-test-observe-adjust loops with explicit hypotheses, evidence, outcomes, and next actions.
SkillBFreshness: BClaude Code +9 - 0.529score
tdd-workflow
projectamazonph
Drive one observable software behavior at a time through a valid Test-Driven Development cycle: select the test, prove Red for the intended reason, implement…
SkillBFreshness: BClaude Code +9 - 0.529score
solid
projectamazonph
Use this skill when writing code, implementing features, refactoring, planning architecture, designing systems, reviewing code, or debugging.
SkillBFreshness: BClaude Code +9 - 0.529score
engineering-quality-gates
projectamazonph
Verify engineering work against required behavioral, architectural, regression, build, security, compatibility, documentation, and diff-hygiene evidence…
SkillBFreshness: BClaude Code +9 - 0.529score
solid-code-design
projectamazonph
Design, refactor, and review object-oriented or modular code with the SOLID principles while preserving behavior and avoiding needless abstraction.
SkillBFreshness: BClaude Code +9 - 0.523score
precision-workflow
projectamazonph
execute complex work with research, files, and quality gates
SkillBFreshness: BClaude Code +9
Cite this page
engineering-standards-orchestrator skill. Agent Mart. https://agentmart.dev/skills/projectamazonph-engineering-standards-orchestrator (last updated 2026-10-24)