From Graph Sync Agent Kit
Custom postprocessor authoring workflow for WordLift graph-sync projects. Use when creating, editing, testing, or debugging graph-sync postprocessors, postprocessor manifests, SDK runner invocations, context payload usage, or postprocessor failure artifacts.
How this skill is triggered — by the user, by Claude, or both
Slash command
/graph-sync-agent-kit:graph-sync-postprocessor-authoringThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Use this skill to create, modify, test, or debug custom postprocessors for WordLift graph-sync projects.
Use this skill to create, modify, test, or debug custom postprocessors for WordLift graph-sync projects.
Read references/runtime.md for runtime loading, manifest behavior, context fields, and failure artifacts. Read references/authoring.md for the class contract, minimal example, and local runner workflow.
When creating or editing a postprocessor:
WebPage identity behavior.When editing manifests:
_base and profile manifest behavior from references/runtime.md._base postprocessors before profile postprocessors.class = "package.module:ClassName".timeout_seconds, enabled, and keep_temp_on_error intentionally.Use the SDK runner for local validation when possible. Resolve the Python runner before executing it:
uv run python -m ... when the project uses uv..venv/bin/python -m ... when the project has a local virtual environment.python -m ... only after confirming it can import wordlift_sdk.python -m wordlift_sdk.kg_build.postprocessor_runner \
--class my_project.postprocessors:AddSimpleNameFallback \
--input-graph ./tmp/input_graph.nq \
--output-graph ./tmp/output_graph.nq \
--context ./tmp/context.json
Inspect:
output/postprocessor_debug/ when keep_temp_on_error = true.For SDK 5.1.1+:
context.settings; read config from context.profile.context.account.key; use context.account_key.context.account as the clean /me account payload.npx claudepluginhub wordlift/agent-marketplace --plugin graph-sync-agent-kitGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.