一键导入
engram-plan-feature
Plan feature implementation by creating an engram task hierarchy. Not a markdown plan — a queryable task tree any agent can retrieve.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Plan feature implementation by creating an engram task hierarchy. Not a markdown plan — a queryable task tree any agent can retrieve.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Document architectural decisions as engraph ADRs. Store context, options considered, rationale, and consequences for queryable decision history.
Every commit must reference a valid engram task UUID. Covers format, validation, and why --no-verify is prohibited.
Orchestrator agent loop. Coordinates work across subagents using engram as shared state. Search before acting, dispatch by UUID, validate before finishing.
Universal session end protocol. Run at the end of every agent session — closes open tasks, generates a handoff summary, syncs to remote, and validates state.
Universal session start protocol. Run at the beginning of every agent session — syncs from remote, opens a session, loads prior context, and surfaces the next action.
Subagent registration protocol. Use when assigned a task UUID by an orchestrator — claim the task, pull context from engram, store all findings, and report results back.
| name | engram-plan-feature |
| description | Plan feature implementation by creating an engram task hierarchy. Not a markdown plan — a queryable task tree any agent can retrieve. |
Create structured plans as engram task hierarchies — not markdown files. Any agent retrieves them via engram_task_show <UUID>.
Per-turn write rule: Every bash, edit, or write call must be followed by
engram_context_createorengram_reasoning_create+engram_relationship_createin the same turn. The planning work below produces context entities at each step — store them immediately, not at the end.
engram_ask query="<feature name> plan"
engram_task_create title="<Feature Name>" priority="high"
# PARENT_UUID
engram_task_update id=<PARENT_UUID> status="in_progress"
For each stage or component:
engram_task_create title="<Feature> Stage N: <Stage Name>" parent=<PARENT_UUID> priority="medium"
# STAGE_UUID
engram_relationship_create source_id=<PARENT_UUID> source_type=task target_id=<STAGE_UUID> target_type=task relationship_type=depends_on agent="<name>"
Store detailed stage instructions as linked context:
engram_context_create title="Stage N detail: <name>" content="<detailed instructions>" source="<reference>"
engram_relationship_create source_id=<STAGE_UUID> source_type=task target_id=<DETAIL_UUID> target_type=context relationship_type=relates_to agent="<name>"
engram_relationship_connected entity_id=<PARENT_UUID> max_depth=2
engram_validate
engram_next
engram_task_show + engram_relationship_connected