メニュー
Spec-driven development with OpenSpec. Use when the user mentions OpenSpec, OPSX, spec-driven development, SDD, change proposals, delta specs, /opsx commands, or wants structured planning for features, migrations, refactors, or bug fixes.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
| name | openspec |
| description | Spec-driven development with OpenSpec. Use when the user mentions OpenSpec, OPSX, spec-driven development, SDD, change proposals, delta specs, /opsx commands, or wants structured planning for features, migrations, refactors, or bug fixes. |
| metadata | {"category":"workflow-automation","pattern":"pipeline","owner":"engineering","status":"active","review_cycle_days":90,"benchmark_after_model_update":true,"outputs":["change-proposal","delta-spec","design","task-list","implementation-verification"]} |
| compatibility | Portable Agent Skills core; OpenSpec CLI is optional but preferred when installed. |
Use OpenSpec as the planning layer when work needs durable requirements, reviewable intent, or a multi-session implementation trail. Keep the repository's normal engineering rules active: start implementation with tests where applicable, work in small steps, and run the project's canonical checks.
openspec/ directory or docs/openspec/.openspec --version.openspec/specs/*/spec.md and active openspec/changes/<change>/ artifacts before proposing edits. If openspec/ is a symlink to docs/openspec/, keep using openspec/... paths for CLI compatibility while recognizing that the physical files live under docs/openspec/.This tooling repo exposes a portable installer through the shared ~/.agents symlink:
~/.agents/skills/openspec/scripts/install-openspec
~/.agents/skills/openspec/scripts/install-openspec /path/to/repo
The installer uses this placement policy:
docs/, create docs/openspec/ and a root openspec -> docs/openspec symlink so openspec CLI commands work from the repo root.thoughts/ but no docs/, create thoughts/openspec/ and a root openspec -> thoughts/openspec symlink.docs/ or thoughts/, create the normal root openspec/ directory.openspec/ directory automatically; treat it as an existing installation.Use the core OPSX path for most changes:
proposal.md, delta specs, design.md, and tasks.md.tasks.md.If slash commands are available in the active tool, prefer:
/opsx:explore for investigation before committing to a change./opsx:propose <change> for the default proposal-to-tasks path./opsx:apply <change> for implementation./opsx:verify <change> when the expanded workflow is enabled./opsx:sync <change> to merge delta specs into canonical specs when needed./opsx:archive <change> to finalize completed work.If the tool does not expose slash commands, perform the same workflow directly in files under openspec/changes/<change>/. In docs-first repositories, that path may resolve through the root symlink to docs/openspec/changes/<change>/.
Prefer CLI commands for repository state and validation when available:
openspec list
openspec list --specs
openspec show <change-or-spec>
openspec status <change>
openspec validate --all
Use --json for agent-readable output when the command supports it. Do not build long-lived automation on beta workspace commands unless the user explicitly asks for experimental workspace setup.
update or wip.SHALL statements and concrete scenarios.GIVEN, WHEN, THEN, and optional AND form.OpenSpec works incrementally. Do not try to generate a full specification library upfront unless the user asks for a documentation pass. For mature codebases, create or update specs around the change being made, then let the spec library grow through real work.
When initializing OpenSpec, choose the smallest tool scope that matches the user's request:
~/.agents/skills/openspec/scripts/install-openspec
openspec update
After upgrading the CLI, run openspec update in initialized repositories so generated tool instructions stay current.
Route durable personal context and reusable knowledge between augmentedcode-configuration and the personal knowledge vault. Use when the user asks to remember, persist, capture, retrieve, or organize personal knowledge; when deciding whether information belongs in rules, skills, portfolio, notes, maps, or inbox; or when using personal-knowledge-vault as a low-context reference layer.
Use when running git clone, git remote add, git remote set-url, or suggesting GitHub SSH commands to ensure all GitHub access uses the saski SSH host alias.
Build modern data apps, dashboards, and interactive reports using either React + Vite or Streamlit. Includes optional Gemini Data Analytics chat integration for an AI powered "chat with your data" experience. Relevant when any of the following conditions are true: 1. User explicitly requests to build a data dashboard, data application, or visualization UI, and the UI pulls data from a GCP database (defaulting to BigQuery unless otherwise specified). 2. You need to generate a frontend web application to interact with, query, and visualize data from GCP data sources. 3. User wants to build a "chat with your data" experience or integrate the Gemini Data Analytics chat API into a web interface. Do NOT use when any of the following conditions are true: 1. The request is for building backend-only services. 2. The request is for simple CLI scripts or command-line applications. 3. The web application is not data-centric or does not involve visualizing/querying data from GCP sources.
Develops and executes Spark code on Dataproc Clusters and Serverless. Reads and writes data using BigLake Iceberg catalogs, BigQuery and Spanner. Debugs execution failures. Use when: - Writing Spark ETL pipelines on GCP. - Training or running inference with ML models with spark on GCP. - Managing Spark clusters, jobs, batches, and interactive sessions. Don't use when: - Writing generic Python scripts that don't use Spark. - Performing simple SQL queries that can be done directly in BigQuery.
Use when the user wants to clean up AI slop, harden a codebase, or run technical due diligence: review code for security flaws and vulnerable or wrongly-licensed dependencies, find and remove dead code, audit dependency health, run and triage linters, type checkers, test suites and formatters, evaluate design principles (SOLID and the cohesion/coupling equivalents for non-OO stacks), audit logging and observability, find performance smells, check documentation and CI health, scaffold baseline configs for the detected stack, do a structured code review, guide a one-major-at-a-time dependency migration, fix project structure, or produce one aggregated health report across all of these. Triggers on phrases like clean this up, find security issues, remove dead code, check dependencies, lint, typecheck, run the tests, format, review my code, check the design, check logging, is this slow, migrate a dependency, fix the project structure, set up the project, audit everything, due diligence, or this looks like AI slop
Run any question, idea, or decision through a council of 5 AI advisors who independently analyze it, peer-review each other anonymously, and synthesize a final verdict. Based on Karpathy's LLM Council methodology. MANDATORY TRIGGERS: 'council this', 'run the council', 'war room this', 'pressure-test this', 'stress-test this', 'debate this'. STRONG TRIGGERS (use when combined with a real decision or tradeoff): 'should I X or Y', 'which option', 'what would you do', 'is this the right move', 'validate this', 'get multiple perspectives', 'I can't decide', 'I'm torn between'. Do NOT trigger on simple yes/no questions, factual lookups, or casual 'should I' without a meaningful tradeoff (e.g. 'should I use markdown' is not a council question). DO trigger when the user presents a genuine decision with stakes, multiple options, and context that suggests they want it pressure-tested from multiple angles.