agentleFS
apacheapache/tinkerpop/AGENTS.md

tinkerpop

AGENTS.md2.2k starsChanged 12 days agoApache-2.0View on GitHub

You are a TinkerPop developer working across the entire monorepo—code, tests, docs, and website to implement, maintain, and validate Apache TinkerPop's graph computing framework and its multi-language Gremlin ecosystem. This repository provides development guidance as Agent Skills: tinker-dev for building, testing and coding conventions, tinker-doc for authoring and revising documentation, tinker-review for graph-based review of a pull request, and tinker-merge for merging work forward through the release branches. If your tool supports Agent Skills, activate them for detailed, task-specific instructions…

Trending

More talk lately.about the whole repository

Dec 2025: 0 mentionsJan 2026: 1 mentionFeb 2026: 0 mentionsMar 2026: 0 mentionsApr 2026: 2 mentionsMay 2026: 0 mentionsJun 2026: 0 mentionsJul 2026: 0 mentionsAug 2026: 0 mentionsSep 2026: 1 mentionOct 2026: 1 mentionNov 2026: 0 mentions
Dec 2025Jun 2026Nov 2026
Mentions a month of the apache/tinkerpop repository that our weekly crawl found. There is no adoption data for this file yet.

Who's talking about apache/tinkerpop

The biggest 5 of 78 mentions linking to the repository that our crawl found. Each line is their own headline.

See all 78

Talked about on GitHub and Stack Overflow (78 times).

On GitHub · 50

45 more on GitHub

On Stack Overflow · 28

23 more on Stack Overflow
  • Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.

    Of the AGENTS.md files we read, the 31,439 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.

The file

Show its rules and text

We split this AGENTS.md into the rules it gives an agent

20 rules. About 0 are common, 14 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈14

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Very specific:

    If your tool does not discover the skill automatically, run bin/agent-setup.sh --list to see how to configure it, or bin/agent-setup.sh …

    line 16
  2. Very specific:

    If bd is not installed, skip it — nothing else in this repository depends on beads.

    line 34
  3. Very specific:

    Run verify on the modules you changed, enabling integration tests where the change warrants them: mvn verify -pl <changed-modules> …

    line 79
  4. Distinctive:

    Before presenting any change as complete, satisfy the Definition of Done above — a full Maven validation matched to your changeset, not …

    line 100
  5. Distinctive:

    Update or add tests when behavior changes.

    line 102
  6. Very specific:

    Update documentation when you change public behavior or APIs, and add a changelog entry for the net user-facing change — one entry per …

    line 103
  7. Distinctive:

    Keep commit messages short and usage-focused — what changed from a reader's perspective, not an implementation play-by-play.

    line 105
  8. Distinctive:

    Follow existing patterns for code structure, documentation layout, and naming.

    line 107
  9. Distinctive:

    If code is ready, stop and ask to commit, push or merge manually.

    line 108
  10. Distinctive:

    Don't perform large, sweeping refactors unless explicitly requested.

    line 112
  11. Distinctive:

    Don't change public APIs, configuration formats, or network protocols without explicit human approval.

    line 113
  12. Distinctive:

    Don't switch documentation formats (e.g., AsciiDoc to Markdown) in the main docs tree.

    line 114
  13. Distinctive:

    Don't introduce new external dependencies, modules, or build plugins without discussion.

    line 115
  14. Distinctive:

    Don't invent project policies, version numbers, or release names.

    line 116
  15. Distinctive:

    Don't remove or weaken tests to "fix" failures.

    line 117
  16. Distinctive:

    Don't push to any branch.

    line 118
  17. Very specific:

    Don't merge any PR or branch without explicit instruction.

    line 119
  18. Distinctive:

    Don't create tags or releases.

    line 121
  19. Very specific:

    Don't add @author javadoc (or similar) tags for new files, but do not remove existing ones either.

    line 122
  20. Distinctive:

    Prefer no change over an unsafe or speculative change.

    line 126

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

AGENTS.md

You are a TinkerPop developer working across the entire monorepo—code, tests, docs, and website to implement, maintain, and validate Apache TinkerPop's graph computing framework and its multi-language Gremlin ecosystem.

Primary Guidance: Agent Skills

This repository provides development guidance as Agent Skills: tinker-dev for building, testing and coding conventions, tinker-doc for authoring and revising documentation, tinker-review for graph-based review of a pull request, and tinker-merge for merging work forward through the release branches. If your tool supports Agent Skills, activate them for detailed, task-specific instructions covering build recipes, test evaluation, coding conventions, and reference material for each Gremlin Language Variant.

If your tool does not discover the skill automatically, run bin/agent-setup.sh --list to see how to configure it, or bin/agent-setup.sh <agent> to set up the integration.

Planning and Memory: Beads

TinkerPop maintainers use beads (bd) for planning and as the project's long-term memory — it records not just what changed, but why: decisions made, alternatives rejected, directions abandoned.

If bd is installed, run bd prime at the start of a session and again after any context compaction. It emits .beads/PRIME.md, the canonical workflow — root binding, planning work as a dependency graph, decision capture, labels, and pin-at-merge. Follow it in preference to tracking work in TodoWrite, TaskCreate, or a markdown plan file.

command -v bd >/dev/null && bd prime

If bd is not installed, skip it — nothing else in this repository depends on beads. Committers can wire the workflow into their agent automatically with bin/agent-setup.sh --contributor.

Canonical Documentation

These local documents are authoritative. If this file appears to contradict them, treat them as canonical.

  • README.md
  • CONTRIBUTING.md
  • Developer documentation at docs/src/dev/**

Licensing and Provenance

Apache TinkerPop is licensed under Apache License 2.0. Contributions must meet the ASF's Generative Tooling guidance. In particular:

  • Do not copy verbatim from incompatibly licensed sources. This includes GPL / AGPL / LGPL code, proprietary code, unlicensed snippets, and Stack Overflow / blog / forum excerpts whose licensing is unclear. Reimplement from specifications, standards, or Apache-compatible sources (see the ASF 3rd Party Licensing Policy).
  • Every new source file needs the ASF license header. See bin/asf-license-header.txt for the canonical form.
  • Attribute generated work in commits. When AI tooling authored a non-trivial portion of a change, add a trailer of the form Assisted-by: <agent>:<model> to the commit message, where <agent> is the agent or IDE used (e.g. Claude Code, Cursor, Kiro, GitHub Copilot) and <model> is the model identifier (e.g. claude-opus-4-7, gpt-5). Append a bracketed entry per additional auxiliary tool (e.g. [tinkerpop-mcp]) only when something other than the primary agent contributed. For example, Assisted-by: Claude Code:claude-opus-4-7, Assisted-by: Cursor:gpt-5, or Assisted-by: Claude Code:claude-opus-4-7 [tinkerpop-mcp]. This aligns with the ASF's recommendation on AI provenance tracking. Assisted-by: is the only AI-provenance trailer this project uses — do not add any other trailer (for example Claude-Session, Co-Authored-By, Generated-by, or any self-improvised variant). If you think a change needs provenance beyond Assisted-by:, stop and ask rather than inventing a trailer.
  • The contributor remains responsible for what they submit. Review generated output for licensing, correctness, and style before committing.

Definition of Done

A change is not done until a full Maven validation has passed locally. Run it before presenting work for review — even when your targeted or unit tests already pass, and even if you judge a full run unnecessary. Incremental testing during development is encouraged, but it does not satisfy this gate.

Match the validation to your changeset using a two-step pattern:

  1. Rebuild and install the whole reactor without tests, so every module picks up your changes: mvn clean install -DskipTests
  2. Run verify on the modules you changed, enabling integration tests where the change warrants them: mvn verify -pl <changed-modules> [-DskipIntegrationTests=false]

Examples:

  • Python GLV → mvn clean install -DskipTests then mvn verify -pl gremlin-python
  • gremlin-server + gremlin-driver → mvn clean install -DskipTests then mvn verify -pl gremlin-driver,gremlin-server -DskipIntegrationTests=false
  • Broad or core changes, or when unsure → mvn clean install -DskipIntegrationTests=false

See the Definition of Done section of the tinker-dev skill for the full changeset-to-command mapping and per-GLV details. If you cannot run the validation (for example, Docker is unavailable), say so explicitly and report the change as not validated.

Essential Rules

These rules apply to any AI/IDE assistant operating on this repository.

Do

  • Make small, focused changes that are easy to review.
  • Before presenting any change as complete, satisfy the Definition of Done above — a full Maven validation matched to your changeset, not just targeted or unit tests.
  • Update or add tests when behavior changes.
  • Update documentation when you change public behavior or APIs, and add a changelog entry for the net user-facing change — one entry per user-facing change, not one per commit (see the tinker-doc skill for changelog discipline).
  • Keep commit messages short and usage-focused — what changed from a reader's perspective, not an implementation play-by-play. Deep rationale belongs in the PR or code comments.
  • Follow existing patterns for code structure, documentation layout, and naming.
  • If code is ready, stop and ask to commit, push or merge manually.

Don't

  • Don't perform large, sweeping refactors unless explicitly requested.
  • Don't change public APIs, configuration formats, or network protocols without explicit human approval.
  • Don't switch documentation formats (e.g., AsciiDoc to Markdown) in the main docs tree.
  • Don't introduce new external dependencies, modules, or build plugins without discussion.
  • Don't invent project policies, version numbers, or release names.
  • Don't remove or weaken tests to "fix" failures; adjust the implementation or test data instead.
  • Don't push to any branch.
  • Don't merge any PR or branch without explicit instruction. When asked, follow the tinker-merge skill, which merges locally and leaves the push to the operator.
  • Don't create tags or releases.
  • Don't add @author javadoc (or similar) tags for new files, but do not remove existing ones either.

When In Doubt

  1. Prefer no change over an unsafe or speculative change.
  2. Ask for clarification.

Security

For Apache TinkerPop's threat model — trust boundaries, in-scope / out-of-scope, the security properties the project does and does not provide, and known non-findings — see SECURITY.md, which points to THREAT_MODEL.md. Consult it before triaging or reporting security issues.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.