Free SKILL.md scraped from GitHub. Clone the repo or copy the file directly into your Claude Code skills directory.
npx versuz@latest install specgraph-specgraph-skills-specgraph-graph-querygit clone https://github.com/specgraph/specgraph.gitcp specgraph/SKILL.MD ~/.claude/skills/specgraph-specgraph-skills-specgraph-graph-query/SKILL.md---
name: specgraph-graph-query
description: Use when the user wants to find specs, see what's ready to work on, trace dependencies, or assess the impact of changing a spec. Routes to spec, graph_query, and the relevant MCP resources.
license: Apache-2.0
metadata:
source: https://github.com/specgraph/specgraph
---
# SpecGraph Graph Query
SpecGraph stores specs as graph nodes with first-class edges (dependencies,
blocks, compositions, decisions). This skill covers reading the graph: finding
specs, listing what's ready, walking dependencies, and impact analysis.
## When this skill applies
Trigger on questions like:
- "What specs are ready for me?"
- "What does X depend on?" / "What depends on X?"
- "Show me the graph" / "What's the impact of changing Y?"
- "List all specs" / "Find specs about Z"
## Tools and resources
| Intent | Use |
|---|---|
| List specs (with filters) | `spec` tool, `action: "list"` |
| Get a single spec | `spec` tool, `action: "get"`, with `slug` |
| Walk dependencies | `graph_query` tool, with `action: "deps"` or `"impact"` |
| Quick "what can I work on" | `specgraph://graph/ready` resource |
| Whole-graph view | `specgraph://graph` resource |
| Spec body and history | `specgraph://spec/<slug>` resource |
## Filters on `spec list`
The `spec` list action accepts:
- `stage` — filter by funnel stage (`spark`, `shape`, ..., `approved`, `done`)
- `domain` — filter by domain tag if the project uses domains
- `text_query` — substring match against intent and slug
Combine filters as needed. A typical "what's in flight" call is
`stage=specify,decompose,approved` plus the project's primary domain.
## Graph traversal
`graph_query` actions:
- `deps` — transitive dependencies of a slug (what it needs)
- `impact` — transitive dependents (what would be affected by change)
- `path` — shortest dependency path between two slugs
All traversals are bounded to 50 hops. Cycles are detected and reported.
## Resource cheat sheet
```text
specgraph://specs # list of all specs
specgraph://spec/{slug} # one spec, full body
specgraph://spec/{slug}/changes # change history
specgraph://decision/{slug} # decision node
specgraph://constitution # full constitution
specgraph://constitution/{layer} # one layer (User/Org/Project/Domain)
specgraph://graph # graph summary
specgraph://graph/ready # specs unblocked and ready
specgraph://findings # all open findings
specgraph://prime # session-priming digest
```
## Don't
- Don't query the graph by reading raw storage paths. Use the MCP tools and
resources; they enforce auth, edge semantics, and pagination.
- Don't infer "ready" from stage alone. A spec is ready when its dependencies
are satisfied — `specgraph://graph/ready` does the math.