AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified Apache-2.0 Self-run

Sa Architecture Design

skill-evolplus-talos-sa-architecture-design · by evolplus

SA design-mode procedure for producing docs/architecture.md, ADRs, and docs/instrumentation-contract.md from a signed-off SRS. Use when Solution Architect is dispatched in `design` mode after SRS sign-off to create or revise greenfield architecture, choose dependencies through the approval gate, define C4 C1-C3, API/data/async/cross-cutting sections, and establish format-boundary and gate-field c…

No reviews yet
0 installs
12 views
0.0% view→install

Install

$ agentstack add skill-evolplus-talos-sa-architecture-design

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access Used
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-evolplus-talos-sa-architecture-design)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Sa Architecture Design? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

SA Architecture Design

Use

Use this skill when SA is dispatched in design mode. Do not use for brownfield extraction or external-integration adequacy; those have separate skills.

Inputs

  • docs/SRS.md at Status: Signed-off
  • docs/user-stories/, docs/frs/, docs/external-integrations/
  • Existing docs/architecture.md and docs/decisions/ when revising
  • SRS headers Designated Dependency Approver, Frontend-Framework, Backend-Track, Backend-Framework
  • solution-defaults, third-party-dependency-evaluation, adr-author, c4-author, data-lifecycle-contracts, format-boundary-contracts

Outputs

  1. docs/architecture.md using agents/_templates/_artifacts/architecture-template.md.
  2. ADRs under docs/decisions/ for every non-trivial choice.
  3. docs/instrumentation-contract.md for any SRS with a UI surface.
  4. docs/open-issues.md entries for architectural gaps that cannot be resolved from the SRS.
  5. Worktree plan-update.json with track: "sa".

Procedure

  1. Verify SRS is signed off. Halt if not.
  2. Read the SRS, every linked US/FR, external integration files, and prior ADRs.
  3. Build C4 levels:
  • C1 Context: system boundary, actors, external systems.
  • C2 Containers: runnable/deployable units, datastores, queues, clients.
  • C3 Components: required for non-trivial containers; each component row carries Linked FRs.
  • Do not author C4 Code; Dev roles own code-level artifacts.
  1. Fill architecture sections:
  • §3.5 Data Models when persistent state exists.
  • §3.6 API Inventory for REST/gRPC/GraphQL/WebSocket/events/queues, linking contract files.
  • §3.7 Dependency & Call Graph for cross-container calls, internal packages, and production dependencies.
  • §3.8 Async Workflows for event/queue flows.
  • Cross-cutting concerns, failure modes, NFR posture, open risks, and observability.
  1. Apply data-contract skills:
  • Use format-boundary-contracts for fields crossing incompatible formats.
  • Use data-lifecycle-contracts for gate fields read by one component and written by another.
  1. For UI-bearing SRSs, author docs/instrumentation-contract.md declaring stable test IDs/accessibility identifiers for FE Dev and QA.
  2. For every non-trivial architectural choice, author an ADR.
  3. For new third-party dependencies:
  • Check solution-defaults first.
  • Reuse prior ADR approvals when valid.
  • Otherwise evaluate 2-4 options with third-party-dependency-evaluation, halt with NEEDS_CONTEXT, and wait for the Designated Dependency Approver.
  • Do not include the dependency in accepted architecture until approval is recorded.
  1. Run a self-containment check on produced artifacts.
  2. Commit changes, then emit plan-update.json.

Quality Bar

The architecture must let TL break down tasks and Devs implement without reverse-engineering intent. Every SRS requirement must have an architectural home, or a documented open issue.

Hard Rules

  • Never propose architecture before SRS sign-off.
  • Never write source code, tests, API contracts, or plan files.
  • Never weaken SRS Security & Compliance.
  • Never approve dependencies yourself.
  • Never skip docs/instrumentation-contract.md for UI-bearing SRSs.
  • Never omit architecture §6 when format-boundary or gate-field conditions are present.
  • Commit before signaling done.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.