genlayerlabs

Interactive wizard to set up a GenLayer validator node on Linux.

Write production-quality GenLayer intelligent contracts. Always pins concrete GenVM runner version hashes and never uses local-only test/latest runner aliases. Covers equivalence principles, storage rules, LLM resilience, and cross-contract interaction.

Write and run fast direct mode tests for GenLayer intelligent contracts.

Use the GenLayer CLI to deploy, interact with, and debug intelligent contracts.

Validate GenLayer intelligent contracts with the GenVM linter.

Write and run integration tests against a GenLayer environment.

Manage GenLayer validators across testnets using the genlayer CLI. Join, fund, set identity, list, and organize validators per network and owner.

Execute git commit with conventional commit message analysis, intelligent staging, and message generation. Use when user asks to commit changes or mentions "/commit".

Create, list, deactivate, and reactivate API keys for rate limiting

Create, list, update, and delete API rate limiting tiers

Fetch the most urgent Linear issue with tag "studio" and size XS, then fix it

Fetch, analyze, fix Sentry issues, run tests, and create PRs

Setup Python virtual environment and run integration tests with gltest

Monitor Discord community channel for user-reported bugs and issues

Debug GenLayer Studio deployments via ArgoCD CLI

Debug and manage Cloudflare DNS, cache, and proxy settings using flarectl

Design, deploy, and monitor a GenLayer Intelligent Oracle prediction market from any coding agent. Use when a user wants to create a settled-by-web-evidence prediction market without opening the web UI.

Deploy GenLayer intelligent contracts using the deployment script. User-invoked only.

Write and run fast direct-mode pytest tests for this repo's GenLayer intelligent contracts.

Use the GenLayer CLI to inspect, interact with, and debug this repo's intelligent contracts.

Validate this repo's GenLayer intelligent contracts with the GenVM linter.

Run or design live-environment GenLayer tests for this intelligent-oracle repo.

Run tests and build checks across the intelligent-oracle project. Use after making changes to validate nothing is broken.

Write production-quality GenLayer intelligent contracts for this intelligent-oracle repo.

GenVM commit message conventions. Use when writing a commit message, squashing/merging a PR, or amending history. Covers the `type(scope): summary <emoji>` format, the five types, the standard scope set, the gitmoji suffix, and mistakes to avoid (typos, vague "fix CI N").

Builds the GenVM project. Use after making code changes to compile Rust binaries.

Read this BEFORE trying to fix a GenVM build that fails on macOS. GenVM is NOT built natively on macOS — building from source on Darwin breaks the deterministic runner-artifact invariant. Use a remote Linux nix-builder instead. Triggers on any build/nix/runner/linker/ar failure on a Mac (Apple Silicon or Intel).

Add RST docstrings to public Python methods/classes. Use when asked to document Python code.

GenVM Rust test conventions and style. Use when writing, adding, or modifying Rust tests — inline `#[cfg(test)] mod tests`, integration tests under a crate's `tests/`, helpers, assertions, and how `ya-test-runner` discovers them.

Runs tests for the GenVM project. Use after making code changes to verify correctness.

Sets up the development environment for GenVM repository. Use when setting up the repo for the first time or when dependencies need to be refreshed.

Wire the Genswarms.Cron scheduler object into a swarm: jobs as datetime + stamped message to an allowlisted target, trusted-source gating, injectable persistence, occurrence-scoped retry + a consecutive-failure breaker. Use when adding scheduled/proactive behavior to a swarm or debugging "job not created" (untrusted source), "job not delivered" (target not allowlisted), or jobs lost on restart (no store_mod).

Wire the Genswarms.Browser object: web browser for agents with two modes (allowlist fail-closed, or denylist allow-any with IP-filtering proxy). Render/click/type/back with compact replies that keep pages OUT of agent context. Use when giving agents web access, or debugging containment issues. Denylist mode requires a deployment-provided IP-filtering egress proxy for sub-resource SSRF containment.

Wire the Genswarms.Tips rotating-content dispenser: fragment pools assembled per recipient (no-repeat rotating slots + weighted dressing slots), seeded deterministic draw/commit, pending->live->retired lifecycle, injectable store. Use when adding rotating outbound copy (daily tips, onboarding nudges, rotating campaign angles) to a swarm, or debugging "empty_pool" replies or repeated content after a restart (no store).

Wire the Genswarms.Metrics counters object: fire-and-forget bumps from other objects, closed key allowlist, periodic flush to an injectable store. Use when adding operational counters to a swarm or debugging "bump ignored" (key not in allowlist) or counters resetting on restart (no store).

Reviews the current GenVM branch by fanning out three specialized review agents (spec, security, implementation). Use when asked to "review this branch", "review the PR", or do a full code review of a diff.

Builds the GenVM project. Use after making code changes to compile Rust binaries.

GenVM Rust test conventions and style. Use when writing, adding, or modifying Rust tests — inline `#[cfg(test)] mod tests`, integration tests under a crate's `tests/`, helpers, assertions, and how `genvm-tool test` discovers them.

Runs tests for the GenVM project. Use after making code changes to verify correctness.

Author Σ_pol policies and Σ_flow flows for this LLM policy host. Load this file into any assistant and it can generate valid `policy_ir` / `flow_ir` terms (as JSON) to POST against the host's OpenAI-compatible endpoint — the live model/provider catalog is embedded below so the policies target models this host actually serves.

Develop and extend the unhardcoded router/host — add a provider (declaration + source + adapter + knobs), touch the OpenAI-compatible shim or the /x/* operator surface, the Σ_flow runner, the Postgres host store, the ingress + dashboard, or the lupa embedding of the engine core. Load this when changing host code, adding a provider or api_kind, editing the catalog, or running the test suite. To change the *policy algebra* itself, that lives in the core submodule — use unhardcoded-engine-contribute. To merely use the router, use unhardcoded-use.

Use a running unhardcoded router — as a CALLER (send an OpenAI-compatible request carrying a Σ_pol policy_ir / Σ_flow flow_ir, or a profile/family/pin model string; session pinning; streaming; dry-run before spending) and as an OPERATOR (dashboard, /x/* endpoints, mint keys, hot-add providers, AntSeed wallet). Load this when integrating an app or agent against the router, wiring a client, minting keys, debugging a route, or operating the host. To AUTHOR a policy term against this host's live catalog use the root SKILL.md (sigma-policy-author); to develop the host itself use unhardcoded-contribute.

Operate and orchestrate genswarms agent swarms — author swarm configs (.exs/.json/.yaml), build the CLI, and start/manage/observe/scale swarms via the genswarms CLI or REST API. Use when working in a genswarms repo, defining or running a swarm, wiring agent topologies/backends/skills, or driving swarms over HTTP/WebSocket. Do NOT use for developing the genswarms Elixir codebase itself, or for unrelated multi-agent frameworks.

Develop and extend genswarms — the Elixir/OTP swarm orchestrator. Load this before adding a backend or an object, or touching the IR control plane, the OTP supervision tree, the messaging layer, the daemon/SQLite coordination, or the Phoenix API. Orients and points into AGENTS.md, CLAUDE.md and docs/. To merely operate a swarm, use genswarms-use.

Drive genswarms — the Elixir/OTP orchestrator for swarms of subzeroclaw agents. Load this to define a swarm (agents + objects + topology), run it as a daemon, and observe/scale/message it via the `genswarms` CLI or the REST + WebSocket API. Orients and points into the root SKILL.md (operating-genswarms) and docs/ for depth. To develop or extend genswarms itself (add a backend or object, touch the IR/API/daemon), use genswarms-contribute.

Develop SubZeroClaw — the single-file ~550-line C agentic runtime. Load this before changing src/subzeroclaw.c or src/test.c: it carries the anti-framework thesis (the goal is NOT to grow), the code map (the loop, config, the shell tool, async compaction), what will and won't be merged, and the build/test loop. To merely run or configure SubZeroClaw, use subzeroclaw-use.

Run and operate SubZeroClaw — the ~550-line C agentic daemon (skill.md + LLM + shell + loop). Load this to build it, configure it (~/.subzeroclaw/config), write an agent skill, run a task, wire it to an unhardcoded router for routing/cache/compaction, or run it as a systemd service with credential isolation. WARNING: it executes arbitrary shell with no sandbox. To develop the runtime itself, use subzeroclaw-contribute.

Extend the unhardcoded-engine (Σ_pol) policy algebra — add a filter predicate, a scorer, a selector, an Xform, a FailPlan action, or a catalog field; and do it without breaking the invariants that make the core trustworthy (determinism, append-only signature, canonical normal form, golden conformance). Load this skill when changing anything under llm_policy/, adding an operator, or touching term admission/normalization/encoding. To merely use or embed the engine, use unhardcoded-engine-use instead.

Use the unhardcoded-engine (Σ_pol) policy algebra — either to EMBED the pure Lua core in a host, or to AUTHOR a policy_ir term that decides which (provider, model) serves an LLM call. Load this skill when you need to integrate llm_policy into a host (lupa/mlua/luerl), write or dry-run a Σ_pol policy, understand the request/trace contract, or debug why a candidate was rejected. For extending the algebra itself (new operators/fields) use unhardcoded-engine-contribute instead.

Develop the gsp CLI — the single static Go binary for content-addressed GenSwarms packages. Load this before changing anything under cli/: the IR package (a second implementation of the genswarms IR that must stay in conformance), the dirhash digest (byte-compatible with the swarmidx server), the notary client, or adding a command. To merely use gsp, use gsp-use.

Author, publish, resolve and verify GenSwarms packages with the gsp CLI against a swarmidx notary. Use when sharing or consuming GenSwarms components (bodies, policies, handlers, agents, swarms), publishing a package to a notary, resolving a swarmidx: ref to a content digest, or verifying a transparency log. Also covers the offline authoring plane (dirhash, materialize/fold overlays, validate IR) that never touches a live swarm.

Cut a release of genlayer-js. Bumps version, updates CHANGELOG, tags, pushes — CI then publishes to npm and creates the GitHub Release. Use when a human asks "release v1.x.y" or "ship a new version".