Apache TinkerPop is a graph-computing framework for applications that store or analyze connected data, such as property graphs in databases. Developers use its Gremlin traversal language, drivers, servers, and tools to work with different graph systems through a common interface.
Borrowing it
Nothing to install: this file belongs to apache/tinkerpop. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/apache/tinkerpop/master/.skills/tinker-review/SKILL.mdgit clone --depth 1 https://github.com/apache/tinkerpopWrote 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/tinkerpop/tinker-review)<a href="https://agentmods.dev/skills/apache/tinkerpop/tinker-review"><img src="https://agentmods.dev/badge/skills/apache/tinkerpop/tinker-review/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/tinkerpop/tinker-review"><img src="https://agentmods.dev/badge/skills/apache/tinkerpop/tinker-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 3 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 281 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 281 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
- high Tool Misuse · line 281 Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).Fix: Validate all tool parameters against an allowlist. Reject dangerous parameter values (shell=True, --force, -rf /) and use safe defaults.
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 | $0.00059 | $0.04854 |
| Opus 5.5 | $0.00024 | $0.01942 |
| Sonnet 5.5 | $0.00012 | $0.00971 |
| Haiku 4.5 | $0.00006 | $0.00485 |
Grade C, and why
tinker-review scanned grade C with 1 finding 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 7d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
(`rm -rf /tmp/pr-review-<pr>`). Re-running the review on the same PR clears it, How it starts
The opening of the file, as written. The whole thing — 341 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Graph Review Skill
Prerequisites
- Docker running (for Gremlin Server)
upstreamremote pointing to[email protected]:apache/tinkerpop.git(fetch only)- Node.js 20+ with dependencies installed in
.skills/tinker-review/ - Beads (
bd) installed, with the repo's beads database available (bin/agent-setup.sh --contributor). The review stops before fetching the PR whenbdis missing. It runsbd dolt pullfirst and falls back to local state if the pull fails; it never writes to beads.
References (load on demand)
- Read references/schema.md when you need to understand what vertices, edges, or properties exist in the knowledge graph (typically during enrichment or when writing raw Gremlin)
- Read references/interfaces.md when you need the exact function signatures or data type definitions for a module
- Read references/enrichment-cli.md when you need to know what an enrichment CLI command does and when to reach for it (the command names below are terse; this is where their meaning lives)
- Read references/functional-testing.md when you run the optional functional test (step 4) — what
functional/cli.jsdoes, how the built server is configured, and how to drive it from the subagent
Execution Sequence
When invoked with /review <pr-number>:
The run has two phases (see DESIGN.md). Phase 1 (step 1) is deterministic and builds the graph. Phase 2 (steps 3–5) is agent-driven — enrichment, an optional functional test, then the report. Step 2 chooses the playbooks that guide Phase 2; step 6 tears down.
1. Setup + Phase 1 (deterministic)
Run the review script. This handles everything mechanical:
npm install --prefix .skills/tinker-review # only needed once
node .skills/tinker-review/scripts/review.js <pr-number> <repo-path>
This performs: check for beads and pull it → fetch PR → create worktree → start Gremlin Server → extract structure via Tree-sitter → populate knowledge graph → discover discussions (JIRA, dev list, proposals, PR comments) → discover the PR's beads (records naming the PR or those discussions, their roots and subtrees) → run pattern checks (completeness, coverage gaps, centrality, blast radius, cluster analysis) → write evidence JSON.
What ships with it
56 files 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.
- .gitignore 34 B
- DESIGN.md 7.6 KB
- package-lock.json 11 KB
- package.json 509 B
- playbooks/bug-fix.md 2.3 KB
- playbooks/driver-server.md 3.7 KB
- playbooks/general.md 8.5 KB
- playbooks/glv.md 2.8 KB
- playbooks/grammar.md 2.2 KB
- playbooks/new-step.md 3.0 KB
- playbooks/removal.md 3.2 KB
- playbooks/tinkergraph.md 6.5 KB
- references/enrichment-cli.md 9.8 KB
- references/functional-testing.md 6.9 KB
- references/interfaces.md 16 KB
- references/schema.md 13 KB
- scripts/discovery/beads.js 13 KB runs code
- scripts/discovery/discussions.js 16 KB runs code
- scripts/enrichment/api.js 25 KB runs code
- scripts/enrichment/cli.js 11 KB runs code
- scripts/extraction/code-index.js 4.0 KB runs code
- scripts/extraction/tree-sitter.js 36 KB runs code
- scripts/functional/cli.js 3.5 KB runs code
- scripts/functional/setup.js 13 KB runs code
- scripts/graph/change-levels.js 6.4 KB runs code
- scripts/graph/confidence.js 3.0 KB runs code
- scripts/graph/externals.js 3.7 KB runs code
- scripts/graph/populate-beads.js 5.1 KB runs code
- scripts/graph/populate-discussions.js 11 KB runs code
- scripts/graph/populate.js 15 KB runs code
- scripts/graph/references.js 3.1 KB runs code
- scripts/infrastructure/docker.js 3.2 KB runs code
- scripts/infrastructure/net.js 2.6 KB runs code
- scripts/patterns/architecture.js 5.5 KB runs code
- scripts/patterns/blast-radius.js 6.9 KB runs code
- scripts/patterns/centrality.js 6.1 KB runs code
- scripts/patterns/classify-externals.js 3.4 KB runs code
- scripts/patterns/cluster-analysis.js 3.0 KB runs code
- scripts/patterns/community-detection.js 20 KB runs code
- scripts/patterns/completeness.js 3.3 KB runs code
- scripts/patterns/confidence-audit.js 6.8 KB runs code
- scripts/patterns/coverage-gaps.js 3.3 KB runs code
- scripts/patterns/orphans.js 3.3 KB runs code
- scripts/patterns/removal-refs.js 4.5 KB runs code
- scripts/renderer/code-links.js 5.7 KB runs code
- scripts/renderer/render.js 42 KB runs code
- scripts/renderer/template.html 8.0 KB
- scripts/review.js 31 KB runs code
- test/beads.test.js 15 KB runs code
- test/cli-docs.test.js 4.4 KB runs code
- test/code-links.test.js 8.0 KB runs code
- test/community-detection.test.js 8.6 KB runs code
- test/discovery.test.js 10 KB runs code
- test/extraction.test.js 8.1 KB runs code
- test/fingerprint.test.js 6.6 KB runs code
- test/playbook-sections.test.js 4.0 KB runs code
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.
- 7d ago Changed · +1 lines b4ea30398dc0
- 19d ago Changed · +13 lines 8ba2f2e3f492
- 20d ago Changed · +29 lines 1d64dd986831
- 2mo ago First seen · 298 lines · 59 tokens per session scan C 3de0fd3f9986
tinker-review is a skill published in the GitHub repository apache/tinkerpop (2,150 stars, last pushed today), licensed Apache-2.0. It adds 59 tokens to every session and 4,854 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
memgraph-storage-reviewer
Review code changes in Memgraph's storage layer, including MVCC, concurrency patterns, WAL, recovery, DDL operations, index/constraint management, delta chains, and skip list operations. Invoke for pull requests or changes to src/storage/v2/.
pr-comment-evaluator
Use to judge whether a PR review comment sounds polite.
diff-risk-explainer
Use to briefly explain small code diffs.
code-graph
This skill should be used when understanding code structure, finding dependencies between functions/classes, tracing call graphs, or exploring code relationships. Trigger phrases include 'code graph', 'call graph', 'who calls', 'what calls', 'find dependencies', 'code structure', 'inheritance', 'find paths'.
data-integrity-review
Data integrity specialist for schema correctness, owner isolation, projections, hash chains, log separation, snapshots, and replay consistency. Use when validating that data remains correct under sync and persistence rules.
architect-review
use when user ask to review spec.