メニュー
Use for project infra and architecture context: environments, adapters, runtime configuration, deployments, service dependencies, datastores, API contracts, ownership, incidents, and dependency blast radius.
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
Use when an agent needs recent or historical change context: what changed recently, regressions, merged PRs, tickets, docs, incidents, deployments, releases, and source-history ingestion.
Use when the task is centered on running, explaining, configuring, or troubleshooting the `potpie` command: doctor, login, pot management, source registration, search, graph workbench reads/writes, and pot scope behavior.
Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts, CI failures, local dev setup issues, repeated bugs, prior fixes, failed attempts, and verification history.
Use when the task can read or write the project-memory graph through the potpie CLI: discover the contract with `graph catalog`, read named views with `graph read`, resolve entity identity with `graph search-entities`, create validated plans with `graph propose`, commit plans with `graph commit --verify`, inspect quality with `graph quality`, or capture uncertain work with `graph inbox`. Also covers writing retrieval-grade descriptions and responding to nudges. Prefer this over the MCP context_* tools whenever the shell is available.
Use before writing, modifying, reviewing, refactoring, or testing code so repo/project preferences surface: error handling, file structure, frameworks, logging, dependency choices, testing, security, API style, and naming. Also use after code work when a reusable project preference should be recorded.
Use when establishing, refreshing, or deeply understanding a repository's baseline memory in Potpie: purpose, application type, features, services/modules, environments, deploy shape, dependencies, API contracts, datastores, integrations, ownership, and explicit preferences. The harness reads authored and code-adjacent sources, then writes graph workbench mutations.
| name | potpie-infra-architecture |
| description | Use for project infra and architecture context: environments, adapters, runtime configuration, deployments, service dependencies, datastores, API contracts, ownership, incidents, and dependency blast radius. |
Use this skill when the task touches environments, adapters, deployment behavior, runtime config, service dependencies, production incidents, or architecture changes.
Start from the service, environment, adapter, or dependency named by the task. Resolve identity before assuming keys:
potpie graph search-entities "<service env adapter dependency>" --limit 10
Read the service neighborhood with the graph workbench:
potpie graph read \
--subgraph infra_topology \
--view service_neighborhood \
--scope service:<service-name> \
--environment <dev|staging|prod|preview> \
--depth 2 \
--direction both \
--limit 20
Omit --environment only when the task is environment-agnostic. For broad
architecture work, run potpie --json graph describe infra_topology --view service_neighborhood --examples before choosing the read shape.
Use infra context for blast radius, where-to-look decisions, deployment changes, adapter selection, and incident debugging. Preserve environment qualifiers; a staging dependency is not proof of a production dependency.
Look for explicit topology facts: DEFINED_IN, DEPLOYED_TO, DEPENDS_ON,
USES, EXPOSES, HOSTED_ON, and OWNED_BY.
Record only source-backed topology or carefully labeled agent inferences. Use
authoritative_fact when evidence is an explicit source of truth such as
deployment config, service manifest, infra doc, ADR, or user statement. Use
agent_claim for lower-authority interpretation.
Use the workbench write flow:
potpie --json graph catalog --task "record infra architecture"
potpie graph search-entities "<service>" --type Service --limit 10
potpie --json graph describe infra_topology --view service_neighborhood --examples
potpie --json graph propose --file mutation.json
potpie --json graph commit <plan_id> --verify
potpie --json graph history --plan <plan_id>
Every durable infra fact needs an environment when the fact differs by environment, evidence when available, and a retrieval-grade description.
Architecture capture is harness-led: inspect authoritative sources and write semantic facts. Do not use scanner-driven graph updates or infer topology from directory names, imports, or package files alone.
Use this only when the potpie CLI is unavailable:
{"intent":"operations","include":["infra_topology","timeline","owners"],"mode":"balanced","source_policy":"summary"}
Include families belong to the envelope surface (context_* MCP tools and
potpie resolve/potpie search); in the graph workbench they are served by the
graph views infra_topology.service_neighborhood, recent_changes.timeline,
and code_topology.ownership_by_path — graph read does not accept include
family names.