This skill is the EffectCraft skill suite'ssole execution implementation: task recovery ledger, isolated Python preparation, native command sessions, review and revision all converge in this directory scripts/ within.effectcraft-use、effectcraft-cli and the various domain skills are thin documentation skills — they provide scenario guidance, plan examples, and reference contracts; actual execution always goes through this skill's entry points.Before executing any EffectCraft command, first set HARNESS_DIR to this SKILL.md directory (the absolute path actually loaded by the host); do not assume the current working directory is the skill directory. User-level installations may be located at ~/.agents/skills/effectcraft-harness, project-level may be located at .agents/skills/effectcraft-harness, plugins may be located at their skills/effectcraft-harness or the host cache directory; the path actually loaded by the host shall prevail.
Six Entry Points
| Entry Point | Purpose |
|---|---|
scripts/launch.sh |
POSIX managed entry point (recommended). Automatically prepares the fixed Python and EffectCraft within the user directory without modifying the global PATH; covers task actions such as plan / run / inspect / resume / review / revise / cancel. |
scripts/launch.ps1 |
Windows PowerShell managed entry point, with the same semantics as launch.sh . |
scripts/cli.py |
Direct CLI channel when isolated Python is ready (for diagnostics and compatibility purposes; does not bypass the task ledger contract). |
scripts/commands.py |
Native command single-skill entry point:list_commands / execute_command / argument directory and real-time pre-checks. |
scripts/workflow.py |
Native composite plan executor, reused internally by launch.sh/launch.ps1 workflow mode; generally not called directly. |
scripts/bootstrap.py |
First-use installer: prepares the locked isolated Python 3.13 and downloads and verifies the official EffectCraft artifact, returning JSON executable。 |
Usage
By default, creative writing uses the managed entry point (subject to the scenario guidance in domain skills such as effectcraft-use ; plan and example files are taken from the corresponding domain skill directory):
HARNESS_DIR=/absolute/path/to/effectcraft-harness
/bin/sh "$HARNESS_DIR/scripts/launch.sh" plan --mode workflow --plan PLAN.json --output /absolute/new-result
/bin/sh "$HARNESS_DIR/scripts/launch.sh" run --mode workflow --plan PLAN.json --output /absolute/new-result
Windows PowerShell：& "$HARNESS_DIR/scripts/launch.ps1" run --mode workflow --plan PLAN.json --output "C:\Users\me\result".
For native command plans, append --mode commands; for your own desktop session, use --mode desktop。
First-time installation (no preinstalled Python; the installation scope is the user data directory and does not require sudo):
python3 -I -B "$HARNESS_DIR/scripts/bootstrap.py"
Native command direct connection and command directory query (native subcommands of cli.py must be placed after -- ; commands.py always outputs JSON):
python3 -I -B "$HARNESS_DIR/scripts/cli.py" -- info --json
python3 -I -B "$HARNESS_DIR/scripts/commands.py" list --filter <关键词>
Relationship Between Domain Skills and This Harness
effectcraft-use: complete task routing and delivery contract; execution still goes through this harness.effectcraft-cliand the various scenario skills (project/footage/composition/layers/animation/effects/masks/expressions/camera/export/tracking/puppet/setup): scenario knowledge and parameter contracts; execution still goes through this harness.- In domain skill documentation,
SKILL_DIRrefers to the domain skill's own directory (used to read itsexamples/、references/); script execution paths are always replaced with this directory'sHARNESS_DIR/scripts/。 - For the execution contract (task recovery ledger, review, revision, resource budget), see
effectcraft-useskill'sreferences/managed-execution.md。