Remote Agent Scheduler

Use when the user wants to create, inspect, update, or run scheduled remote agents in the cloud rather than local cron jobs.

Security Scan

The skill's instructions match its stated goal (scheduling remote agent runs) but are ambiguous about authentication and credential handling—there's no declared credential or connector mechanism, which could lead to unsafe collection of secrets or unclear cloud access behavior.

Purpose & Capability

Name, description, and instructions align: the skill is about creating/listing/updating/running scheduled remote agents and asks for repo, environment, connectors, schedule, and prompt payload — all relevant. However, provenance is minimal (no homepage, unknown source) and the SKILL.md doesn't specify which cloud provider or management API it will use.

Instruction Scope

SKILL.md stays within the scheduling domain (build/inspect trigger definitions, surface auth and repo blockers). It explicitly asks the agent to "confirm authentication, environment, and connector assumptions," but does not limit how authentication is obtained or handled — this could lead the agent to request or log secrets in-chat or in other places.

Install Mechanism

Instruction-only skill with no install spec and no code files to run. Lowest install risk (nothing is written to disk by the skill itself).

Credentials

No required environment variables or declared primary credential are listed, yet the skill needs repo access, environment/connector context, and to confirm authentication. The absence of a declared credential mechanism is an inconsistency: the skill will likely require credentials (cloud API keys, repo tokens, connector OAuth) but provides no safe or explicit way to supply them. That gap increases the risk that users will be asked to paste sensitive tokens into chat or elsewhere.

Persistence & Privilege

always:false and no install hooks; the skill does not request persistent agent-wide privileges or claim to modify other skills or system settings.

What to consider before installing

This skill looks like it does what it says (schedule remote agent runs) but is vague about how it will authenticate to repos or cloud services. Before installing or using it: (1) ask the author which cloud provider or scheduler API it uses and where triggers are stored; (2) do NOT paste API keys, tokens, or passwords into chat — insist on OAuth-based connectors or a secure secrets mechanism; (3) require a clear, minimal-scope credential model (which env vars or connector tokens are needed and what permissions they require); (4) if you plan to let it perform actions, request the full source or an explicit integration endpoint (e.g., docs for the cloud scheduler API) so you can verify behavior; (5) prefer ephemeral or least-privilege tokens and audit logging for scheduled runs. If the author cannot explain how credentials are provided securely, treat the skill as unsafe to use.

Use this skill for recurring remote agent jobs that run in isolated cloud environments.

Workflow

1. Determine whether the user wants to create, list, update, or run a trigger.
2. Collect repo source, environment, connectors, schedule, and prompt payload.
3. Build or inspect the trigger definition.
4. Confirm authentication, environment, and connector assumptions.
5. Return the trigger details and next management steps.

Guardrails

• Distinguish remote cloud runs from local cron jobs.
• Keep repo URL, environment ID, and schedule explicit.
• Surface setup blockers like missing auth or repo access early.

Example Requests

• Create a nightly remote agent to summarize repo health.
• Update this cloud trigger to use a different environment and cron.

Inputs

• Desired action
• Repo source
• Environment and connector context
• Cron schedule

Outputs

• Trigger definition
• Run or update confirmation
• Operational notes

Success Criteria

• The action type is clear.
• Remote environment assumptions are explicit.
• Auth and repo blockers are surfaced.

Non-Goals

• Local cron setup
• Ambiguous cloud-vs-local scheduling

Source Provenance

Derived from src/skills/bundled/scheduleRemoteAgents.ts.