Skill

plan-new

Drafts implementation plan from ready PRD and research artifacts after passing PRD/research gates. Starts plan stage in feature-dev-aidd workflow.

Python

Bash

developer-tools

automation

npx claudepluginhub grinrus/ai_driven_dev --plugin feature-dev-aidd

Configuration

Model: inherit

Tool Access

This skill is limited to using the following tools:

ReadEditWriteGlobBash(rg *)Bash(sed *)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/plan-new/runtime/research_check.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 *)

Preview

Follow `feature-dev-aidd:aidd-core`.

Supporting Assets

runtime/research_check.pytemplates/plan.template.md

SKILL.md

Similar Skills

plan

Creates requirements chain (idea → PRD → plan) for large features or initiatives, or single plan documents for focused work. Use for 'plan X', 'roadmap for Y' requests.

archcore

plan-phase

Generates execution-ready implementation plans from research docs in .artifacts/research/. Ensures junior developers can execute tasks with zero ambiguity. Planning only—no code changes, fixes, or verification.

10 tools

harness-engineering

writing-plans

Transforms research findings into actionable implementation plans with granular tasks, verification criteria, and stakes classification. Use before coding to ensure alignment.

rpikit

Stats

Stars5

Forks0

Last CommitApr 17, 2026

Actions

View Source View Plugin View on GitHub View README

Help us improve

Share bugs, ideas, or general feedback.

Follow feature-dev-aidd:aidd-core.

Steps

Set active stage plan and active feature.
Run python3 ${CLAUDE_PLUGIN_ROOT}/skills/plan-new/runtime/research_check.py --ticket <ticket>.
Gate readiness with python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/prd_check.py and python3 ${CLAUDE_PLUGIN_ROOT}/skills/plan-new/runtime/research_check.py; block early if either fails.
Use the existing rolling context pack as input evidence; do not invoke standalone context-pack builder scripts from this stage.
Runtime-path safety: execute only canonical runtime commands from this contract (python3 ${CLAUDE_PLUGIN_ROOT}/skills/.../runtime/...).
Root-relative /skills/... runtime paths are forbidden; use only ${CLAUDE_PLUGIN_ROOT}/skills/.../runtime/....
Run subagent feature-dev-aidd:planner. The planner is the only writer for aidd/docs/plan/<ticket>.md.
Run subagent feature-dev-aidd:validator. The validator is read-only and returns the narrative verdict and gap list only.
Final stage readiness is sourced from prd_check + research_check + validator verdict; return /feature-dev-aidd:review-spec <ticket> only on the ready path.
Otherwise return PENDING or BLOCKED with the validator gaps and the canonical next action for the current stage.

Command contracts

`python3 ${CLAUDE_PLUGIN_ROOT}/skills/plan-new/runtime/research_check.py`

When to run: before planning to enforce research-readiness gate.
Inputs: --ticket <ticket> and current PRD/research artifacts.
Outputs: deterministic readiness verdict for plan stage.
Failure mode: non-zero exit if research status/artifacts do not satisfy gate policy.
Next action: resolve research blockers and rerun the gate check.

`python3 ${CLAUDE_PLUGIN_ROOT}/skills/aidd-flow-state/runtime/prd_check.py`

When to run: before planner execution and before final stage verdict normalization.
Inputs: active PRD artifacts for the current ticket.
Outputs: deterministic PRD readiness verdict used by the stage command as source-of-truth input.
Failure mode: non-zero exit when required PRD fields or readiness markers are missing.
Next action: fix the PRD blockers, then rerun the gate before planning continues.

Notes

Planning stage: AIDD:ACTIONS_LOG: n/a.

Additional resources

Plan template source: templates/plan.template.md (when: generating or normalizing plan structure; why: ensure planner/validator outputs follow canonical sections).