potpie-graph
Use when the task can read or write the project-memory graph through the potpie CLI: discover…
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.
$ npx -y skills add potpie-ai/potpie --skill potpie-change-timeline --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/potpie-change-timelineContext preview
The summary Claude sees to decide when to auto-load this skill.
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.
name: potpie-change-timeline description: "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 this skill when the user asks what changed, when debugging a possible regression, or when ingesting source history from GitHub, Linear, Jira, docs, or deployment records.
Read the project timeline first. A pot is the project boundary and can contain multiple repos, so do not narrow to the current repo unless the user asks.
potpie graph read \ --subgraph recent_changes \ --view timeline \ --format table \ --time-window 7d \ --limit 20
Use the user's exact dates when provided:
potpie graph read \ --subgraph recent_changes \ --view timeline \ --format table \ --since 2026-06-01 \ --until 2026-06-15 \ --limit 50
Only narrow when the user gives a service, environment, or topic:
potpie graph read \ --subgraph recent_changes \ --view timeline \ --format table \ --scope service:<service-name> \ --query "<symptom feature deployment>" \ --time-window 14d \ --limit 20
Timeline context is correlation, not proof. Use it to choose files, PRs, tickets, or deploys to inspect, then verify the source ref before blaming a change. Timeline reads do not include uncommitted local work unless it was recorded.
For GitHub, Linear, Jira, docs, and similar sources, hydrate records with the agent's integration tools/connectors first. Do not use Potpie CLI queue ingestion as the source-history path.
Use the workbench write flow after reading the source:
potpie --json graph catalog --task "record timeline change" potpie --json graph describe recent_changes --view timeline --examples potpie --json graph propose --file mutation.json potpie --json graph commit <plan_id> --verify potpie --json graph history --plan <plan_id>
Use the source event time for `occurred_at`, not ingestion time. Add fixes, decisions, bug patterns, or infra links only when the source explicitly supports them.
Timeline capture is harness-led. Do not use scanner-driven graph updates or turn source titles into facts without reading the source.
Use when the task can read or write the project-memory graph through the potpie CLI: discover…
Use while debugging or troubleshooting failures, flaky tests, incidents, production alerts,…
Use for project infra and architecture context: environments, adapters, runtime…
Use before writing, modifying, reviewing, refactoring, or testing code so repo/project…
Use when establishing, refreshing, or deeply understanding a repository's baseline memory in…
Use when the user explicitly asks to ingest, refresh, or deeply understand a repository, PR,…