workflow-architect
Org role guidance for a workflow architect: map full workflow trees for systems, user journeys and agents, covering every branch and failure. Covers discovering workflows from code, handoff contracts, timeouts, recovery paths and observable state.
- License 许可
- Apache-2.0
- 所属仓库
- monoes/monomind
- 最近更新
- 2026-11-10
- Stars
- ★ 23
- Forks
- 2
01 前言 Front Matter
- name
- workflow-architect
- tags
- [ "leadership", "operations", "architecture", "planning" ]
- tools
- []
- source
- https://github.com/monoes/monomind
- license
- Apache-2.0
- description
- Org role guidance for a workflow architect: map full workflow trees for systems, user journeys and agents, covering every branch and failure. Covers discovering workflows from code, handoff contracts, timeouts, recovery paths and observable state.
03 SKILL.md 正文
Workflow Architect — Best Practices
Focus
Maps complete workflow trees for systems, user journeys, and agent interactions — happy paths, every branch, failure modes, recovery paths, and handoff contracts — before implementation starts.
Best practices
- Discover workflows before designing them: read every route, worker, migration, and infra config — most workflows are implied by code, never announced.
- Design branches, not just the happy path: every step needs its validation failures, timeouts, transient/permanent failures, partial failures, and concurrent conflicts covered.
- Define explicit handoff contracts at every system boundary — payload shape, success/failure response, timeout value, and the recovery action on failure.
- Specify observable state at every step: what the customer sees, what the operator sees, what's in the database, what's in the logs.
- Build a cleanup inventory for every resource a workflow creates — every created resource needs a corresponding destroy path on failure.
- Verify specs against actual code, not stated intent — code and description diverge constantly; find and surface the divergence.
- Track every unverifiable assumption explicitly in the spec — an untracked assumption is a future bug.
Common pitfalls
- Specifying only the success case and leaving failure/timeout/partial-failure branches implicit or unwritten.
- Making implementation decisions instead of specifying required behavior — that's the implementer's job, not the spec's.
- Leaving a handoff's failure response or timeout undefined, so downstream code guesses.
- Letting the spec drift silently from the code after either one changes.
Tools & techniques
- Structured workflow-tree spec format: Trigger -> Steps (with timeout/success/failure) -> Cleanup Inventory -> State Transitions -> Test Cases.
- Discovery pass over routes, workers, migrations, and IaC before writing any spec.
- Derive one test case per branch in the tree — an unspecced branch is an untested branch.
- Cross-check every spec with a reality pass against the actual implementation before marking it approved.
04 同仓库更多 Skills
zk-steward
Org role guidance for a ZK steward: engineer zero-knowledge proof systems with circuit design, SNARK vs STARK choice and security review. This is zero-knowledge cryptography, not Zettelkasten note-taking.
writing-plans
Use when you have a spec or requirements for a multi-step coding task and must write an implementation plan before touching code. Produces bite-sized TDD tasks with exact file paths, code and commands for an engineer with zero context.
write-spec
Write a feature spec or PRD from a problem statement or feature idea. Use when turning a vague idea or user request into a structured document, scoping a feature with goals and non-goals, defining success metrics and acceptance criteria, or breaking a big ask into a phased spec.
write-query
Write optimized SQL for your dialect with best practices. Use when translating a natural-language data need into SQL, building a multi-CTE query with joins and aggregations, optimizing a query against a large partitioned table, or getting dialect-specific syntax for Snowflake, BigQuery, Postgres, etc.
workflow-automation
Org role guidance for workflow automation: build and maintain CI/CD definitions in GitHub Actions or equivalent: triggers, jobs, matrices, caching. Covers path filters, dependency caching, parallel jobs, timeouts and fail-fast ordering.
worker-specialist
Org role guidance for a worker specialist: execute one assigned, well-scoped task with precision and report status honestly at real milestones. Covers confirming done-criteria first, surfacing blockers immediately and sharing intermediate results with peers.