From phronesis
Request, read, and verify energy-transition forecasts from the Phronesis platform — grid resilience, demand-load, capacity planning, renewable deployment, and carbon intensity. Use this skill when an agent needs a structured, uncertainty-banded, audit-verifiable forecast about energy systems rather than a free-text guess.
How this skill is triggered — by the user, by Claude, or both
Slash command
/phronesis:v1_energyThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Phronesis V1 (Energy) forecasts the energy transition: load growth, renewable
Phronesis V1 (Energy) forecasts the energy transition: load growth, renewable deployment trajectories, grid stress, and carbon intensity. Use this skill when an agent must ground an energy-related decision in a calibrated forecast with an uncertainty band, explicit assumptions, cited sources, and an audit trail — not a free-text estimate.
https://phronesis-jrstinehour.replit.appPythia-Energy, Themis cluster themis-V1-energyPick the archetype that matches the decision before calling the endpoint:
demand-load-forecast — electricity demand / load growth over a horizon.renewable-deployment — solar / wind / storage build-out trajectories.grid-resilience — grid stress, congestion, and reliability under scenarios.carbon-intensity — grid carbon-intensity trajectory for a region.Confirm the archetype is live by calling GET /v1/catalog and checking the V1
entry's archetypes_available list and status: LIVE.
Call the JWT-gated Decision API:
POST /v1/decision/forecast
Authorization: Bearer <JWT>
Content-Type: application/json
{
"vertical": "V1",
"archetype": "demand-load-forecast",
"subject": "energy",
"question": "Peak summer load for the ERCOT grid region in 2028",
"horizon": "2028",
"region": "ERCOT",
"compute_tier": "deep"
}
compute_tier is abstract — one of standard, deep, strategic, or
strategic-sync. Deeper tiers buy more reasoning depth and source breadth; the
per-call cost is attested in the response, not guessed by the caller.
The response is a contract-v1 envelope: { "status": "ok", "data": { ... }, "request_id": "..." }. The data object is the canonical decision forecast:
p50 value with units.p10 / p50 / p90. The band is monotonic
(p10 <= p50 <= p90); always present this band, never the point alone.audit_trail_id — the id used to fetch the Trust Receipt later.GET /v1/trust/receipt/{forecast_id} returns a public,
signed receipt: the scrubbed forecast, its methodology attestation, and
timestamps. Model identity and algorithm internals are intentionally stripped.GET /calibration/leaderboard shows per-vertical
accuracy (MAPE, RMSE, pinball, Brier) versus naive-LLM and other baselines. Check
the V1 (Energy) row for empirical track record before relying on a forecast.GET /v1/substrate/completeness attests how complete
the V1 data substrate is.p50.strategic compute tier and
cite the Trust Receipt audit_trail_id in your own output so the decision chain
stays verifiable.npx claudepluginhub sustainable-finance-partners/skills --plugin phronesisProvides UI/UX resources: 50+ styles, color palettes, font pairings, guidelines, charts for web/mobile across React, Next.js, Vue, Svelte, Tailwind, React Native, Flutter. Aids planning, building, reviewing interfaces.
Fetches up-to-date documentation from Context7 for libraries and frameworks like React, Next.js, Prisma. Use for setup questions, API references, and code examples.