tasks-new

name	tasks-new
description	Derives or refines tasklist items from PRD and plan artifacts. Use when tasklist stage prepares implementation-ready work items. Do not use when the request is plan authoring in `plan-new` or loop execution in `implement`.
argument-hint	$1 [note...]
lang	en
prompt_version	1.1.25
source_version	1.1.25
allowed-tools	["Read","Edit","Write","Glob","Bash(rg *)","Bash(sed *)","Bash(cat *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/tasks-new/runtime/tasks_new.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/set_active_stage.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/set_active_feature.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/prd_check.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-rlm/runtime/rlm_slice.py *)","Bash(python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/tasklist_check.py *)"]
model	inherit
disable-model-invocation	true
user-invocable	true

Follow feature-dev-aidd:aidd-core.

Steps

1. Inputs: resolve active feature and verify PRD/plan artifacts needed for tasklist generation.
2. Preflight: set active stage tasklist and active feature.
3. Orchestration is bounded: run python3 ${CLAUDE_PLUGIN_ROOT}/skills/tasks-new/runtime/tasks_new.py --ticket <ticket>, then gate PRD readiness with python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/prd_check.py.
4. Use the existing rolling context pack as input evidence; do not invoke standalone context-pack builder scripts from this stage.
5. Run subagent feature-dev-aidd:tasklist-refiner only when upstream prerequisites are already present and the current failure mode is repairable tasklist structure. First action: read the rolling context pack.
6. Postflight: validate via python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/tasklist_check.py.
7. Retry policy: allow at most one bounded retry, and only when validator issues are categorized as repairable_structure. Do not retry for upstream_blocker, env/policy errors, or truth/advisory warnings.
8. Forbidden recovery paths in this stage: creating missing upstream artifacts, launching ad hoc repair interviews, reading runtime source files for self-diagnosis, or looping tasks_new.py -> tasklist_check.py -> manual edits without new evidence.
9. Ready path: return /feature-dev-aidd:implement <ticket> only when tasklist_check.py passes and the next iteration is implementation-ready.
10. Pending or blocked path: return the output contract with structured tasklist gaps and a canonical next action on review-spec or tasks-new; stop after the bounded retry budget is exhausted.

Command contracts

python3 ${CLAUDE_PLUGIN_ROOT}/skills/tasks-new/runtime/tasks_new.py

• When to run: as canonical tasklist stage entrypoint before implement phase.
• Inputs: --ticket <ticket> and current PRD/plan artifacts.
• Outputs: normalized tasklist structure and structured validator issue set.
• Failure mode: non-zero exit when upstream blockers exist or tasklist contract checks fail.
• Next action: rerun only after fixing upstream readiness blockers, or do one bounded retry for repairable_structure.

Notes

• Planning stage: AIDD:ACTIONS_LOG: n/a.

Additional resources

• Tasklist template source: templates/tasklist.template.md (when: generating or normalizing tasklist sections; why: keep iteration/handoff structure aligned with canonical template).