Aiconfig Online Evals
Resolve old bookmarks or prompts that still reference the retired LaunchDarkly skill name and route agents to the current online-evals instructions.
Overview
aiconfig-online-evals is a deprecated Operate-phase alias that tells agents to invoke the online-evals skill instead of running any instructions here.
Install
npx skills add https://github.com/launchdarkly/agent-skills --skill aiconfig-online-evalsWhat is this skill?
- Hard stop: do not execute any instructions in this skill package
- Redirects all callers to the renamed online-evals skill
- Preserves Apache-2.0 metadata for old documentation and automation strings
- Explicit rename path from aiconfig-online-evals to online-evals
- Renamed to online-evals; version metadata 0.1.0
Adoption & trust: 1.3k installs on skills.sh; 16 GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
Your repo or prompt still references aiconfig-online-evals after LaunchDarkly renamed the skill, so agents might follow stale or empty guidance.
Who is it for?
Updating legacy skill slugs in docs, saved Claude/Cursor prompts, or internal runbooks that still say aiconfig-online-evals.
Skip if: Running online evaluations, configuring LaunchDarkly experiments, or any task that needs procedural steps—use online-evals exclusively.
When should I use this skill?
An old reference, saved prompt, or automation still calls aiconfig-online-evals instead of online-evals.
What do I get? / Deliverables
After loading this stub, the agent stops and switches to online-evals so eval work uses the current SKILL.md rather than this redirect.
- Redirect message directing agents to online-evals
Recommended Skills
Journey fit
Production AI evaluation and flag-driven monitoring sit in Operate; this slug is only a legacy alias for that capability. Online evals tie to watching live model and config behavior, which maps to monitoring even though this file performs no eval work itself.
How it compares
A rename pointer in the skills catalog, not a runnable eval workflow or MCP server.
Common Questions / FAQ
Who is aiconfig-online-evals for?
Solo builders and maintainers who inherited old references to the LaunchDarkly eval skill name and need agents to land on the correct online-evals package.
When should I use aiconfig-online-evals?
Only when an automated list or historical link still invokes this slug—then let the redirect fire once and change the reference to online-evals; do not use it during Idea, Build, or Ship planning for new eval setups.
Is aiconfig-online-evals safe to install?
It is a short redirect with no task logic, but review the Security Audits panel on this Prism page and prefer installing online-evals for actual work.
Workflow Chain
Then invoke: online evals
SKILL.md
READMESKILL.md - Aiconfig Online Evals
# Deprecated — renamed to `online-evals` This skill has been renamed. **Do not follow any instructions here and do not perform the task from this skill.** Invoke the **`online-evals`** skill instead — it contains the current, complete instructions. If you arrived here from an old reference (documentation, a saved prompt, or automation), update it from `aiconfig-online-evals` to `online-evals`.