Help us improve
Share bugs, ideas, or general feedback.
From Metis
Produce an implementation plan for a free-text task description. Plan stays in chat.
npx claudepluginhub gsaranti/pantheon --plugin metisHow this skill is triggered — by the user, by Claude, or both
Slash command
/metis:metis-plan-taskThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Produce one implementation plan for a free-text task description. The plan stays in chat; nothing is written to disk.
Guides technical evaluation of code review feedback: read fully, restate for understanding, verify against codebase, respond with reasoning or pushback before implementing.
Share bugs, ideas, or general feedback.
Produce one implementation plan for a free-text task description. The plan stays in chat; nothing is written to disk.
/metis-plan-task "implement Stripe webhook signature verification"..metis/BUILD.md and .metis/INDEX.md (when it exists) are expected to be in context already (loaded by /metis-session-start). BUILD.md is the architecture; INDEX.md is the concept → source-doc map, useful for finding the passages a description points at.
If either is missing — the session began without a session-start — read them lazily as the plan turns on them. Subsequent turns are cache hits.
.metis/INDEX.md to look up source-doc paths for concepts the description names, .metis/BUILD.md to identify likely architecture sections, then read source docs at the named paths.metis-code-explorer subagent dispatches when the surface is unfamiliar enough that planning would be guessing.metis-domain-researcher subagent when a plan step would commit to a technical choice the corpus does not cover..metis/CONTRADICTIONS.md, .metis/QUESTIONS.md, .metis/RESOLVED.md — these belong to the reconcile/walk flow, not to planning.BUILD.md and what the description points at.references/planning-a-task.md — read before drafting.
None. The plan is presented in chat as structured markdown. If the session ends mid-plan or before implementation, /metis-session-end captures the relevant plan content in .metis/CURRENT.md so the next session can resume.
The free-text description after the command is the planning prompt — required, not optional. An empty description means stop and ask the user what to plan rather than guessing.
The plan, in chat, with the shape from references/planning-a-task.md.
After the plan: one line on the next step (/metis-implement-task to execute), and a tip: "For Claude Code's native plan-mode review UI, run this in plan mode (Shift+Tab) before invoking."
If the task appears already substantially done (per the "already done" check in references/planning-a-task.md), return the evidence finding instead of a plan.