hugegraph-ai connects the HugeGraph graph platform with artificial intelligence tools, including large language models, graph retrieval, knowledge-graph construction, and graph machine learning. It is intended for developers building applications that query, analyze, or reason over graph data through Python interfaces and related modules.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add apache/hugegraph-ai --skill hugegraph-regression-testergit clone --depth 1 https://github.com/apache/hugegraph-aiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/apache/hugegraph-ai/hugegraph-regression-tester)<a href="https://agentmods.dev/skills/apache/hugegraph-ai/hugegraph-regression-tester"><img src="https://agentmods.dev/badge/skills/apache/hugegraph-ai/hugegraph-regression-tester/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/apache/hugegraph-ai/hugegraph-regression-tester"><img src="https://agentmods.dev/badge/skills/apache/hugegraph-ai/hugegraph-regression-tester.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00054 | $0.00706 |
| Opus 5.5 | $0.00022 | $0.00282 |
| Sonnet 5.5 | $0.00011 | $0.00141 |
| Haiku 4.5 | $0.00005 | $0.00071 |
Grade A, and why
hugegraph-regression-tester scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 50 lines — stays where its author put it; the contents beside it link to each section on GitHub.
HugeGraph Regression Tester
Tool Routes
| Test goal | Tool |
|---|---|
| MCP, graph status, or schema | inspect_graph_tool(include_raw_schema=false) and when needed inspect_graph_tool(include_raw_schema=true) |
| Structured read | query_graph_data_tool (bounded page or exact ID query) |
| Natural-language Gremlin generation | generate_gremlin_tool(query, execute=false) |
| Generate Gremlin without execution | generate_gremlin_tool(query, execute=false); raw execution must return FEATURE_DISABLED |
| Raw Gremlin disabled boundary | execute_gremlin_read_tool(unsafe_write_query); rejection verifies unconditional disablement, not write classification |
| Text-to-graph extraction | extract_graph_data_tool(text, graph_schema?, example_prompt?) |
| Import dry-run | import_graph_data_tool(mode="ingest", graph_data, dry_run=true) |
| Import preview boundary | Assert confirmable=false, preview_only=true, no plan_id; confirmation returns FEATURE_DISABLED |
| Verify existing graph data | query_graph_data_tool |
| Schema design | design_schema_tool(operations?) |
| Schema validation | apply_schema_tool(mode="validate", operations) |
| Schema dry-run | apply_schema_tool(mode="dry_run", operations) |
| Raw write Gremlin disabled boundary | execute_gremlin_write_tool(gremlin_query); must remain disabled even in admin mode |
| Admin gate check without a write | refresh_vid_embeddings_tool(confirm=false) in separate admin-disabled and admin-enabled processes |
Order
inspect -> generate/read -> extract -> import dry-run -> assert preview-only boundary
-> verify read -> schema validate/dry-run -> safety/gate checks
Structured query and plan lifecycle tools require the default core toolset. For a legacy deployment, enable HUGEGRAPH_MCP_TOOLSET=core and restart before using them.
With HUGEGRAPH_MCP_READONLY=false, the admin-disabled refresh call must return
FEATURE_DISABLED with details requiring HUGEGRAPH_MCP_ADMIN_MODE=true; the
admin-enabled call must return CONFIRM_REQUIRED. Keep confirm=false in both
processes to avoid a refresh. Restart between configuration changes. These calls
do not verify AI-side authentication or target validation. Validate the write
classifier independently with unit tests; raw Gremlin rejection cannot exercise it.
For the refresh contract, run the API/MCP regression tests covering default
GraphSpace equivalence, required fields and write/auth gates on
POST /v1/vid-embeddings/refresh, configuration locking, and mixed-version routing.
The capability GET is only discovery; MCP must never fall back to the legacy POST
when the versioned route is unavailable. This does not verify physical-instance
identity or establish a live rolling-deployment test.
Generation and extraction require HUGEGRAPH_MCP_ALLOW_AI=true and a configured AI service.
These skills describe current source; published 1.7.1 uses v2_core / v1.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago Changed · +15 lines e1aa753a5f1d
- 19d ago First seen · 35 lines · 54 tokens per session scan A b1746566b3df
hugegraph-regression-tester is a skill published in the GitHub repository apache/hugegraph-ai (144 stars, last pushed 3d ago), licensed Apache-2.0. It adds 54 tokens to every session and 706 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-27.
Other skills, from other repositories
schema-evolution-migrator
Keep a production knowledge graph healthy across schema evolution, node/edge lifecycle, incremental updates, and coordinated deployment. Emits N-1 compatible Neo4j-Migrations-style schema migrations, temporal-invalidation Cypher (bitemporal tvalid/tinvalid, invalidate-not-delete), LightRAG-style incremental merge from…
subgraph-access-control
Scope what each agent persona can see in a knowledge graph. Generates Neo4j fine-grained GRANT/DENY policy (traverse on node labels + relationship types, read on properties) per persona role, enforces security transparency (out-of-scope nodes are invisible, not access-denied), handles PII via the UUID-separation…
graduated-validation-protocol
The Ch7 safety envelope for a self-evolving agent: the RPO spine (Recursion, Provenance, Optimization) plus the Graduated Validation Protocol that gates what reaches production. Assigns every candidate change a risk tier and applies the matching scrutiny: Tier 1 canary (1% traffic, automatic rollback), Tier 2 staging…
graph-model-selector
Select a graph data model — labeled property graph (LPG) vs RDF vs hypergraph — by scoring REASONING REQUIREMENTS against five implementation features (formal reasoning, n-ary relations, performance, tool ecosystem, constraint expressiveness), per Ch3 "Evaluating Graph Models". Also models the n-ary -> hyperedge…
review-prs
Review a GitHub pull request in the googleapis/mcp-toolbox repo against the team's reviewer checklist: PR title/description conventions, linked issue, logic errors and unhandled edge cases, breaking changes, test coverage, docs updates, security (input handling), and new dependencies. Use whenever a maintainer asks…
graphify
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a persistent knowledge graph with god nodes, community…