supergraph
Graph evidence uses codebase-memory-mcp scoped by CBM_PROJECT. Orientation
loads get_architecture aspects overview, layers, boundaries, clusters,
and hotspots, plus validated hubs, bridges, and test-gaps recipes. If
optional graph evidence is unavailable, label it unavailable and use Serena or
filesystem evidence.
Dispatch to the right skill. 1% rule: if it might apply, invoke it.
Priority
- User CLAUDE.md / AGENTS.md (highest)
- This skill set
- Default system behavior
Dispatch Table
| User intent |
Skill to invoke |
| "start", "begin", unclear scope |
/supergraph:analyze |
| "plan", "lên kế hoạch", explicit plan request |
/supergraph:plan |
| "implement", "build", "execute" (plan saved) |
/supergraph:execute |
| "fix bug", "debug", "why failing" |
/supergraph:fix |
| "test", "TDD", "add tests" |
/supergraph:tdd |
| "refactor", "clean up", "reorganize" |
/supergraph:plan (then execute) |
| integration/e2e tests, after unit green |
/supergraph:integration |
| "done?", "verify", before commit/PR |
/supergraph:verify |
| "review", "merge", "PR", before merge |
/supergraph:review |
| Need project context |
/supergraph:scan |
Adaptive Gate
- < 20 lines, ≤2 files, no hub/bridge, complexity <10 → use
/supergraph:tdd directly
- ≤5 files, clear requirement, no cross-boundary →
/supergraph:plan with lightweight tasks
- >5 files, ambiguous, hub/bridge, cross-boundary →
/supergraph:analyze first, then plan
User Instructions
"Add X" or "Fix Y" is the WHAT. Skills determine HOW. Never skip a skill because the task "feels simple."
Red Flags (STOP if you catch yourself):
- "I'll just do it quickly" → pick a skill, follow it
- "I already know what to do" → skills evolve, load current one
- "This is too simple for a skill" → simple becomes complex fast
Integration
All skills depend on /supergraph:scan being loaded first in the session.
Subagent Guard
When dispatching subagents (executor, code-reviewer, plan-writer, plan-reviewer):
- Subagents get self-contained prompts — never session history
- Subagents do NOT trigger skills independently — this is the orchestrator's job
- Subagents report results back; orchestrator decides next step (fix, verify, review)
1---2name: supergraph3description: Meta-orchestrator — dispatches to the right skill for the task. Use before any non-trivial work.4---56# supergraph78Graph evidence uses `codebase-memory-mcp` scoped by `CBM_PROJECT`. Orientation9loads `get_architecture` aspects `overview`, `layers`, `boundaries`, `clusters`,10and `hotspots`, plus validated `hubs`, `bridges`, and `test-gaps` recipes. If11optional graph evidence is unavailable, label it unavailable and use Serena or12filesystem evidence.1314Dispatch to the right skill. 1% rule: if it *might* apply, invoke it.1516## Priority17181. User CLAUDE.md / AGENTS.md (highest)192. This skill set203. Default system behavior2122## Dispatch Table2324| User intent | Skill to invoke |25|-------------|----------------|26| "start", "begin", unclear scope | `/supergraph:analyze` |27| "plan", "lên kế hoạch", explicit plan request | `/supergraph:plan` |28| "implement", "build", "execute" (plan saved) | `/supergraph:execute` |29| "fix bug", "debug", "why failing" | `/supergraph:fix` |30| "test", "TDD", "add tests" | `/supergraph:tdd` |31| "refactor", "clean up", "reorganize" | `/supergraph:plan` (then execute) |32| integration/e2e tests, after unit green | `/supergraph:integration` |33| "done?", "verify", before commit/PR | `/supergraph:verify` |34| "review", "merge", "PR", before merge | `/supergraph:review` |35| Need project context | `/supergraph:scan` |3637## Adaptive Gate3839- **< 20 lines, ≤2 files, no hub/bridge, complexity <10** → use `/supergraph:tdd` directly40- **≤5 files, clear requirement, no cross-boundary** → `/supergraph:plan` with lightweight tasks41- **>5 files, ambiguous, hub/bridge, cross-boundary** → `/supergraph:analyze` first, then plan4243## User Instructions4445"Add X" or "Fix Y" is the WHAT. Skills determine HOW. Never skip a skill because the task "feels simple."4647## Red Flags (STOP if you catch yourself):4849- "I'll just do it quickly" → pick a skill, follow it50- "I already know what to do" → skills evolve, load current one51- "This is too simple for a skill" → simple becomes complex fast5253## Integration5455All skills depend on `/supergraph:scan` being loaded first in the session.5657## Subagent Guard5859When dispatching subagents (executor, code-reviewer, plan-writer, plan-reviewer):60- Subagents get self-contained prompts — never session history61- Subagents do NOT trigger skills independently — this is the orchestrator's job62- Subagents report results back; orchestrator decides next step (fix, verify, review)
datit309