| name | map-systems |
| description | Decompose a game concept into systems, map dependencies, and maintain the systems index. |
| argument-hint | [next] |
| user-invocable | true |
Purpose: Turn a game concept into a systems index and a clear design order.
Use when:
design/gdd/game-concept.md exists and the project needs a system map
- you need the next
design-system target from the current MVP surface
Do not use for:
- inventing the game concept itself
- writing individual system GDDs
- broad production-stage readiness assessment
Inputs / Required Context:
- required:
design/gdd/game-concept.md
- optional:
design/gdd/game-pillars.md, existing design/gdd/systems-index.md, existing system docs in design/gdd/*.md, relevant prototype reports in prototypes/*/REPORT.md
Outputs / Owned Artifacts:
- owns
design/gdd/systems-index.md
- creates
design/gdd/ if it does not exist
- uses
gamedev/templates/systems-index.md
- maintains a
High-Risk Systems section inside design/gdd/systems-index.md
Modes or Arguments:
- no argument: create or refresh
design/gdd/systems-index.md
next: recommend the highest-priority next design-system target
Execution Rules:
- Stop immediately if
design/gdd/game-concept.md does not exist.
- Enumerate explicit systems, required inferred systems, and clearly out-of-scope ideas.
- Build dependency order across foundation, core, feature, presentation, and meta layers.
- Assign priority tiers with a bias toward MVP and vertical-slice usefulness.
- Use a fixed status vocabulary for systems and keep updates forward-only:
identified, designed, prototyped, informed-by-prototype, implemented, integrated.
- Apply the status meanings consistently:
identified: the system is only mapped in the systems index
designed: the canonical GDD exists
prototyped: a relevant prototype report exists, but its findings are not fully folded into the canonical design docs
informed-by-prototype: prototype findings and baseline decisions are reflected in the GDD or index
implemented: production code exists for the system
integrated: the implementation is wired into the main playable loop or production flow
- Never downgrade a row when refreshing the index; keep the strongest confirmed status already evidenced.
- Always include a compact
High-Risk Systems section in design/gdd/systems-index.md with up to 5 real MVP risks, and do not pad the list with weak filler. Use:
System
Risk Type
Why It Is Risky
Mitigation
Prototype Candidate
Evidence Needed
- For each high-risk row, name a real closure path:
Prototype Candidate should name a concrete prototype route, or explicitly say why downstream evidence is expected to close the risk without a prototype.
Evidence Needed should name the exact proof that will close the row, not a vague aspiration.
- If prototype reports or durable downstream evidence already exist, use them to update relevant system statuses, risk notes, and
High-risk systems without evidence coverage instead of treating the index as a fresh document.
- Create or update
design/gdd/systems-index.md using the canonical template.
- Preserve the project's working or documentation language and player-facing terminology policy when it is already captured in the concept or technical-preferences docs; do not let the systems index drift into a different default language without an explicit reason.
- In
next mode, refresh the index first if it is missing or stale, then recommend the next highest-leverage move.
- In
next mode, prefer a target that either:
- unlocks multiple downstream systems, or
- reduces a high-risk uncertainty that blocks confident design.
- In
next mode, if the highest-leverage move is evidence gathering rather than spec writing, recommend prototype before design-system.
Failure / Stop Conditions:
- stop if the concept is too vague to separate MVP systems from wish-list ideas
- do not write system GDDs or silently invent missing concept intent
Return Format:
- systems index path
- top MVP systems
- high-risk systems
- unresolved high-risk systems without evidence coverage
- any system statuses updated from existing evidence
- next recommended move:
design-system or prototype
- rationale for that recommendation
Example Invocation:
Related Skills / Boundary:
- hand off to
design-system for one system at a time
- hand off to
prototype when a high-risk system needs validation before deeper design