Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsBlogPro
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Authors
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges
  • Chrome Extension
  • Skill Manager

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

ProTermsPrivacyRefunds
Back to skills

Architect

ASecurity

Ground a nontrivial design, compare distinct caller-first type and module sketches, then implement within scope and redesign when evidence contradicts the shape.

2 stars
0 votes
0 copies
0 views
Added 10/1/2026
ai-agentsgit

Security Analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned 10/6/2026

$npx -y skills add williamwue/oh-my-stack --skill architect --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Architect?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Architect
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/williamwue-architect/badge)](https://www.skillsdirectory.com/skills/williamwue-architect)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
Files
SKILL.md
---
name: architect
description: "Ground a nontrivial design, compare distinct caller-first type and module sketches, then implement within scope and redesign when evidence contradicts the shape."
disable-model-invocation: true
---

# Architect

## Child session handoff

Read [the handoff contract](../poteto-mode/references/subagent-handoff.md).
New tasks, repair rounds, retries, and queue items use fresh child sessions
with the original brief, every later directive, prior findings and responses,
and unresolved objections. Reuse only for required costly live state, and
only when the host allows it. Stop and fence active writers before replacement.
A host-owned orchestrator's model catalog, workspace binding, child tools,
and review-round rules take precedence over the native binding above.
Keep its task handles and attribution receipts. Do not use a backing child
conversation as a new delegated review, or claim native-record verification
for a host-owned child. Report attribution evidence gaps explicitly.

Design before implementation. Track grounding, sketches, synthesis/checkpoint,
implementation, and redesign. A design-only request stops at a design. This
workflow does not expand a request for advice into permission to edit code.

## Ground

Run [how](../how/SKILL.md) on affected subsystems, retaining traced entry points,
data flow, ownership, boundaries, and invariants. For changing ownership or
layering, also run [why](../why/SKILL.md) and carry forward Preserve / Change /
Avoid / Risk constraints with their confidence. Genuinely greenfield work may
skip existing-system grounding, but must still identify integration constraints.

## Sketch and synthesize

Read [the design contract](references/design-contract.md). Run
[arena](../arena/SKILL.md) with the same grounded problem for each runner and
that design contract. Require at least two structurally distinct candidates,
not two cosmetic versions of one design. State this requirement in the common
brief. If candidates converge, ask for a concrete alternative within the budget
or mark alternative exploration incomplete; do not silently waive the requirement.
When configured, pass the ordered `architect.runners` panel to Arena rather
than its general runner panel; its length sets the default candidate count.

Candidates write caller usage first, derive types/signatures and module
ownership from it, and mark bodies as pseudocode or not implemented. Keep
unfinished scaffolding out of the user's working code unless scaffold editing
was requested as part of implementation. Screen red flags before synthesis.
The parent selects and grafts against the rubric, recording interface depth,
invariants, tradeoffs, rejected alternatives, and open questions.

## Checkpoint and implement

Respect an explicit request to stop for approval. Otherwise proceed only when
the user requested implementation and the chosen scope is authorized. A design
request ends with the sketch, rationale, risks, and next implementation step.
Publication, destructive migration, or new external authority still needs its
own authorization.

For implementation, replace placeholders incrementally and verify each usable
unit against the caller examples and invariants. Maintain a deviation ledger:
changed signature or ownership, triggering evidence, and whether the sketch,
requirement, or implementation was wrong. Do not absorb contradictions silently.

## Redesign when needed

Repeated workarounds across unrelated paths, internal details leaking to
callers, type escape hatches, or two independent deviations of the same shape
trigger reassessment. A single legitimate edge case is not automatically a
failed architecture. Re-run how on the actual implementation, make the new
constraints explicit, subtract unnecessary structure, and return to arena
within the agreed budget. Preserve user changes; redesign never authorizes
destructive reset or discarding unrelated work.

Return caller usage, chosen types/signatures and module map, synthesis rationale,
tradeoffs, alternative shapes, open risks, and the next step. If implementation
ran, also report deviations and checks of the final code. Clearly distinguish
a verified design sketch from implemented and behaviorally verified software.

Attribution

williamwuewilliamwue
View sourceSee grades on GitHubMore from williamwue →
SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Ship a skill? Prove it's safe.

Free 120-pattern security scan, letter grade, and an embeddable README badge.

Submit a skill

Related Skills

Caveman

Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens". Stays on until "stop caveman" or "normal mode".

1100021 votes

Hyperplan

Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...

698431 votes

Writing Skills

Create and manage Claude Code skills in HASH repository following Anthropic best practices. Use when creating new skills, modifying skill-rules.json, understanding trigger patterns, working with hooks, debugging skill activation, or implementing progressive disclosure. Covers skill structure, YAML frontmatter, trigger types (keywords, intent patterns), UserPromptSubmit hook, and the 500-line rule. Includes validation and debugging with SKILL_DEBUG. Examples include rust-error-stack, cargo-dep...

3931 votes

Mcp Code Execution

Routes multi-tool workflows through MCP servers for large datasets and pipelines. Use when Bash tool overhead is limiting throughput on data-heavy tasks.

3421 votes

catchup

Recovers the conversation and failed tool calls of a previous Codex, Amp, Claude Code, Antigravity, Cline, Copilot CLI, Cursor, DeepSeek Harness, Grok Build, Kimi, OpenCode, Pi Agent, or ZCode session. Use when the user says "catch up", "what did the last session do", "get me up to speed", "I switched agents", asks to recover/summarize a previous session before continuing, or asks to diagnose or report a catchup failure. Do NOT use for the current conversation, git history, or any non-agent log.

741 votes
View all in ai-agents →