القائمة
Use when development work is large, risky, multi-session, multi-agent, interrupted, or needs durable plan/handoff memory for continuity across coding agents.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Use when a development task is unclear, multi-step, risky, or needs coordinated gates for shared understanding, research, parallel lanes, workflow artifacts, and verification. For a single concern, use the matching gate skill if available; for tiny mechanical changes, skip unless the user asks.
Use when development work has independent lanes for research, codebase exploration, inventory, validation, review, or non-overlapping implementation that can run in subagents, parallel tool calls, or separately owned workstreams.
Use when development work depends on current or external facts: latest docs/APIs, library or platform behavior, security guidance, unfamiliar errors, migrations, comparisons, best practices, prior art, or known fixes.
Use when a development request is vague, exploratory, conflicted, under-specified, or the user says they are unsure, wants to be grilled, wants assumptions challenged, or needs goal, scope, constraints, and success criteria clarified before planning or implementation.
Use before saying development work is done, or when choosing project-shaped tests/checks to prove code behavior, bug fixes, refactors, config, infrastructure, or UI changes.
Pi-only web research workflow for environments that have pi-9router-ext and a subagent extension such as @tintinweb/pi-subagents configured. Use this skill when the user explicitly asks to use 9router, pi-9router-ext, ninerouter tools, or a 9router route/combo for web research, or when project/user instructions state that 9router-web-researcher is the preferred research stack. Do NOT use for ordinary web research in standard shared-mcp-only or non-Pi clients; use web-researcher instead. This skill always delegates research lanes to subagents and then synthesizes compact cited findings.
| name | workflow-artifact |
| description | Use when development work is large, risky, multi-session, multi-agent, interrupted, or needs durable plan/handoff memory for continuity across coding agents. |
Create durable task memory when chat context is not enough.
Default path:
docs/ai-workflow/YYYY-MM-DD-<task-slug>.md
Use an existing repo convention instead when one is obvious.
If the path already exists, update it when it is the same task. For a different task on the same day, append a short distinguishing suffix such as -2 or a more specific slug.
Default to a repository document that is eligible for version control, not a gitignored scratch file, when the artifact is meant to survive agent sessions. Do not commit it unless the user asks or the repo workflow explicitly requires it.
Use a repo-ignored scratch location only when the artifact is local-only or disposable. Mark scratch artifacts clearly and delete them before finishing unless they are still needed for handoff.
When the task completes, set Current Status to Done, Blocked, or Handoff, refresh Verification and Handoff, and either keep the artifact as useful project memory or remove/archive it if it was only live working state.
Create before implementation for large/risky, multi-agent, interrupted, multi-session, or continuity-heavy medium work. Update after meaningful plan changes, consequential decisions, research findings, verification, interruption, or handoff.
Do not create artifacts for tiny work unless the user asks.
# <Task Title>
## Goal
## Mode
Tiny | Small | Medium | Large/Risky
## Current Status
## Local Instructions
Project instruction files consulted and notable constraints.
## Decisions
Consequential decisions only. Include rejected approaches when useful.
## Research
Source links, date-sensitive findings, and implementation impact.
## Plan
Checklist of current work slices.
## Key Areas
Important files, modules, tests, config, or docs. Not a full changelog.
## Verification
Commands, checks, results, manual steps, and gaps.
## Open Questions And Risks
## Handoff
What the next agent should do first.
The artifact is complete only when every template section is present and non-empty or marked N/A with a reason; Current Status, Plan, Verification, Open Questions And Risks, and Handoff match the latest work; and Handoff names the exact next action a fresh agent should take first.