agentsclimarketplace

Hatch3r architecture review

Skill hatch3r/hatch3r/skills/hatch3r-architecture-review

Evaluates architectural decisions and produce ADRs following the project template. Use when making architectural decisions, evaluating trade-offs, or creating ADRs.From its SKILL.md

Install
npx -y skills add hatch3r/hatch3r --skill hatch3r-architecture-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

6.0 KB, as published. Nobody here has run it

Architecture Review Workflow

Quick Start

Task Progress:
- [ ] Step 0: Detect ambiguity (P8 B1)
- [ ] Step 1: Read existing ADRs and the template
- [ ] Step 2: Define the decision context — problem, constraints, options
- [ ] Step 3: Evaluate options — pros/cons, prototype if needed, check ADR constraints
- [ ] Step 4: For external library docs and current best practices, follow the project's tooling hierarchy
- [ ] Step 5: Write ADR following the template
- [ ] Step 6: Update affected specs or docs to reference the new ADR

Step 0 — Detect Ambiguity (P8 B1)

Before any work, scan the invocation for unresolved questions in scope, intent, acceptance criteria, target environment, or irreversibility. If any are found, ask the user via the platform-native question tool per agents/shared/user-question-protocol.md. Do not proceed under silent assumption. Default path, not an exception. Triggers for THIS skill: problem framing (what decision needs to be made), constraint set (mandatory vs preferred), evaluation horizon (short-term vs long-term cost), supersedes which prior ADR, and ADR status target (PROPOSED for discussion vs ACCEPTED for binding decision).

Fan-out Discipline (P8 B2)

Fan-out scales with task size; token cost never justifies serializing independent work (rules/hatch3r-fan-out-discipline.md P8 B2; agents/shared/efficiency-patterns.md). Emit sub_agents_spawned: { count, rationale, task_structure } in your output.

Step 1: Read Existing ADRs and Template

  • Read all ADRs in project docs to understand current architecture and constraints.
  • Read the ADR template (e.g., docs/adr/0001_template.md or equivalent).
  • Template sections: Status, Date, Decision Makers, Context, Decision, Alternatives Considered, Consequences, Compliance, Related.
  • Check if any existing ADR is superseded or conflicts with the proposed decision.
  • Read foundational ADRs for invariants.

Step 2: Define the Decision Context

  • Problem: What issue or opportunity motivates this decision?
  • Constraints: Technical (stack, performance budgets), organizational (timeline, team), regulatory (privacy, security).
  • Options: List 2–4 viable alternatives. Include "do nothing" or "status quo" if relevant.
  • Stakeholders: Who is affected? Who decides?

Reference:

  • Performance budgets: project rules
  • Privacy/security invariants: project rules and specs
  • Data model: project documentation
  • Event model: project documentation

Step 3: Evaluate Options

  • For each option, list pros and cons.
  • Consider: complexity, maintainability, performance, security, cost, migration effort.
  • Prototype or spike if uncertainty is high — document findings.
  • Check against existing ADR constraints: does this align with core architecture?
  • Use a comparison table (as in the template):
OptionProsCons
A......
B......

Step 4: External Research

  • For external library docs and current best practices, follow the project's tooling hierarchy.
  • Issue/PR search (check platform in .hatch3r/hatch.json): Search for related issues, prior discussions, or similar decisions in the repo:
    • GitHub: Use GitHub MCP or gh issue list --search "..." / gh pr list --search "..."
    • Azure DevOps: az boards query --wiql "SELECT [System.Id] FROM WorkItems WHERE [System.Title] CONTAINS '...'" or az repos pr list
    • GitLab: glab issue list --search "..." / glab mr list --search "..."
  • Context7 MCP: Look up current API patterns for relevant libraries.
  • Web search: For novel problems, security advisories, or best practices.

Step 5: Write ADR

Follow the project ADR template:

  • Title: ADR-XXXX: Short descriptive title
  • Status: PROPOSED (draft) or ACCEPTED (approved)
  • Date: YYYY-MM-DD
  • Decision Makers: Names/roles
  • Context: Problem, constraints, motivation
  • Decision: What we are doing
  • Alternatives Considered: Table with pros/cons
  • Consequences: Positive, negative, risks
  • Compliance checklist:
    • Aligns with privacy/security specs
    • Aligns with performance budgets
    • Does not introduce new dependencies without justification
  • Related: Specs, issues, previous ADRs

Save as docs/adr/XXXX_short-title.md (or project convention). Use next available number.

Step 6: Update Affected Docs

  • Add references to the new ADR in relevant specs (e.g., data model, event model, quality engineering).
  • Update ADR index if the project maintains one.
  • Link from related issues/work items or PRs/MRs on the platform.
  • If superseding an ADR, update the old ADR's Status to SUPERSEDED by ADR-XXXX.

Error Handling

  • Conflicting ADRs discovered: If an existing ADR contradicts the proposed decision, reference both ADRs and recommend superseding the older one with explicit justification.
  • Insufficient context for trade-off evaluation: If the codebase lacks enough information to evaluate an option, state the knowledge gap and recommend a time-boxed spike before finalizing the decision.
  • Stakeholder disagreement on recommended option: Document all positions with their rationale and escalate to the project lead with a clear recommendation and dissent summary.

Definition of Done

  • ADR written following project template
  • Options evaluated with pros/cons table
  • Tooling-hierarchy followed for external research where relevant
  • Compliance checklist completed
  • Affected specs or docs updated to reference the ADR
  • No privacy/security invariant violations
  • Performance budgets considered

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.