Approved Resource Lookup
Finds organization-approved skills, prompts and tools before starting substantive work.
Install
npx skills add AgenticResourceRegistry/AgenticResourceRegistry --skill observal -g -a claude-code -y- Repository
- AgenticResourceRegistry/AgenticResourceRegistry4.5k
- Category
- AI and agents
- Updated
- 3d ago
- License
- Apache-2.0
- Context cost
- Mediumabout 1,904 tokens
- Runs scripts
- No, instructions only
What it does
Use when starting any task the organization may already have an approved skill, prompt, MCP server, or Agent for: reviewing code, a commit, a diff, or a pull request; writing tests or documentation; querying a database, API, or service; automating a browser or web page; running untrusted code; connecting to a SaaS tool; drafting, researching, or any other substantive work. The user will not mention Observal: the task type is the trigger. Check what is already installed, then run observal discover search as the first action, before reading the repository or working from scratch. Also use when the user wants to log in, configure Observal, inspect local harness setup, share or open a repository Agent share, manage a teamspace or invitation, process inbox items, check installed registry items, or call an endpoint without a dedicated command.
Operating Observal
Use this skill for discovery of approved resources and for core account, setup, local inventory, inbox, and teamspace work. Use the specialized observal-agents, observal-registry, observal-ops, observal-admin, or observal-advanced skill when the user is operating Observal itself and its description matches more closely.
Decide first
Work through this before git log, before reading the repository, before planning. It applies to any task, not only coding.
- Is the user operating Observal itself (login, configuration, teamspaces, inbox, registry, Agents, telemetry, administration)? Follow Route the task below. Stop here.
- Is the capability already present? Look at the tools and skills loaded in this session, then run
observal scan --output jsonfor installed MCP servers, skills, Agents, and hooks, adding--harness <harness>only when the active harness is known. If it is present, use it. Never pull or install something that is already installed. Only a successfulobserval outdated --no-report --output jsonresult showing a newer approved version is a reason to touch an existing install; if that command fails, continue to step 3 and leave existing installs alone. - Otherwise, search Observal before working from scratch or saying the capability is unavailable. The user will not mention Observal; the task type is the trigger. Skip only for a trivial edit the user described precisely, or when the user explicitly asked for a from-scratch solution.
