Relewise/relewise-agent-plugins / plugins/relewise/skills/relewise-agent-gateway
relewise-agent-gateway
UnverifiedProvides the shared execution layer for Relewise Agent Gateway. Use whenever another Relewise skill needs to discover Datasets or execute an operation through the bundled CLI, MCP, or direct REST, and when diagnosing transport or authentication failures.
Relewise Agent Gateway
Use this skill as the technical execution capability behind the Relewise domain skill that matched the user's request. The domain skill owns operation selection, business rules, interpretation, and mutation safeguards. This skill owns connection bootstrap, transport selection, authentication handling, and invocation mechanics.
Do not replace or weaken instructions from the calling domain skill. Preserve explicit authentication requirements across all fallbacks. For setup or repair with missing MCP tools, perform the MCP connection preflight before treating MCP as unavailable.
Resolve the bundled CLI
Resolve paths from this skill's directory, which is the directory containing this SKILL.md.
- On Windows, try
scripts/relewise-agent.ps1with PowerShell. - On macOS or Linux, try
scripts/relewise-agent. - If the platform launcher is absent, try
relewise-agentand thenrelewise-agent.exefromPATHas appropriate for the host. - Validate the candidate once with
--version, retain its absolute path for the current task, and do not repeat discovery before every call.
Do not recursively search the filesystem, home directory, temporary directories, plugin caches, or other skills. A missing, incompatible, or non-executable launcher means the CLI is locally unavailable; it does not mean Agent Gateway or the requested Dataset capability is unavailable.
Execute an Agent Gateway operation
Read transport selection before the first Agent Gateway call in a task. It defines bootstrap discovery, task-, Dataset-, and operation-scoped state, post-bootstrap policy selection, authentication boundaries, fallbacks, and failure classification. Retain Dataset policies independently: one workflow may require different transports for different Datasets.
When the bundled CLI is selected, read CLI usage. Use the exact REST operation ID provided by the domain skill's operations.json, inspect unfamiliar schemas, and preserve the domain skill's safeguards.
When MCP is selected, use the related tool named by the domain skill's operations.json. A mapping to a REST operation identifies the same product capability but does not make their input or output schemas interchangeable; use the schema exposed by MCP.
When direct REST is selected, resolve current HTTP mechanics from https://my.relewise.com/agents/openapi/v1.json and use base URL https://my.relewise.com/agents/api/v1. Use the exact operation ID provided by the domain skill and the schema exposed by OpenAPI.
For operations exposing pageNumber and pageSize pagination, use one-based pages and pageSize 1 through 100 for records (default 50). Use pageSize: 0 for totals only; pagination.totalPages is then zero. Keep filters stable when requesting subsequent pages and stop at pagination.totalPages. Preserve the selected transport's parameter placement, including nested pagination objects when required. Other limits and cursor-based operations follow their own schemas.
Protect credentials
OAuth-capable MCP clients should use the Agent Gateway Connected App flow. The bundled CLI and the plugin's direct REST fallback currently use PAT authentication; MCP clients without an OAuth-managed Agent Gateway connection may also require a PAT. Prefer an already configured vendor-protected credential or secure credential provider. The portable fallback is RELEWISE_AGENT_GATEWAY_TOKEN.
Never request, print, inspect, log, or persist the token. Never place it in conversation text, command arguments, visible command text, input files, source files, URLs, or output. If user action is required to establish authentication, preserve the calling setup skill's instructions and wait for confirmation.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 469k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 321k | Updated | Claude CodeCodex | Verified | Install |