Stats
Actions
Tags
From idfkit
Use for ALL Python coding tasks that import the idfkit library: creating, parsing, querying, or editing EnergyPlus models in code; validating against the schema; tracking cross-references; building geometry and zoning; standing up HVAC (templates or hand-authored loops); evaluating schedules; computing thermal properties; downloading weather data; running EnergyPlus; and parsing simulation results. Discovers and loads version-matched reference docs from the user's installed idfkit (>=0.13). Triggers: idfkit, import idfkit, IDFObject, epJSON, .idf file, HVACTemplate, AirLoopHVAC, PlantLoop, idfkit.simulation, idfkit.weather, idfkit.schedules, idfkit.thermal, idfkit.visualization, LATEST_VERSION, building energy model in Python.
How this skill is triggered — by the user, by Claude, or both
Slash command
/idfkit:developing-with-idfkitThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
idfkit (>=0.13) ships detailed, task-oriented reference documentation for writing
idfkit (>=0.13) ships detailed, task-oriented reference documentation for writing
Python code against the library inside its own pip package. The bundled skill
is a routing SKILL.md plus a references/ folder of focused topic docs (parsing,
reference tracking, geometry, HVAC, schedules, results, weather, and more), with
code examples that are linted and type-checked so they run as-is.
This skill discovers the idfkit installed in the user's project and loads the reference set that matches that exact version — so the guidance never drifts from the API the code will run against.
Use it when the task is writing or debugging Python code that imports idfkit
(import idfkit, idfkit.simulation, idfkit.weather, …).
This is distinct from interacting with an IDF model through the idfkit MCP
server (the load_model, validate_model, run_simulation, … tools and the
idfkit://… resources). For that tool-driven workflow, use the MCP server and the
workflow skills (simulate, validate, new-model, …) instead. The references
here are about the library API, not the MCP tool surface.
Run the discovery script with the user's project directory:
python ${CLAUDE_SKILL_DIR}/scripts/discover.py --project-dir <USER_PROJECT_DIR>
Use the literal ${CLAUDE_SKILL_DIR} token so the command matches the allowed-tools permission
pattern and runs without a prompt.
The script prints either:
SKILL.md. Read it; it routes into
references/<topic>.md. Fetch the smallest reference that covers the task rather
than pre-loading everything.ERROR: block on stderr (non-zero exit). Follow the printed instructions
(install or upgrade idfkit, activate the right environment) and re-run.${CLAUDE_SKILL_DIR} resolves to the directory containing this file; <USER_PROJECT_DIR> is the
absolute path to the user's project. Passing --project-dir matters because the
script resolves .venv, ../.venv, <git-root>/.venv, Pipfile, poetry.lock,
pdm.lock, and uv.lock relative to it — so it inspects the user's project
environment, not this skill's install location.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub idfkit/idfkit-plugin --plugin idfkit