executing-plans
executing-plans
Use 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
cli-operationsDiscover and operate the LeptonAI lep CLI safely, including command-tree help, aliases, abbreviations, output handling, and read-first mutation gates.
Independent GitHub snapshot · Automated classification · Original documentation governs
Use this sub-skill when the task is about the global lep command line itself: discovering commands, checking the installed CLI version, orienting to command groups, interpreting help output, handling hidden aliases or click abbreviations, scoping logs/output, or planning whether a command is read-only, mutating, or destructive.
Do not use this sub-skill for credential setup details, resource-specific create/update recipes, or Python SDK calls. Route those to the neighboring workspace/auth, workload-management, storage/secrets/ingress, or SDK sub-skills after this sub-skill has established the safe CLI plan.
lep is missing, help is ambiguous, a blank option is rejected, a version warning appears, read-only commands hit auth/network errors, or destructive confirmation is unclear.python scripts/lep_help_snapshot.py --groups endpoint workspace job pod ingress storage file --json
The script runs only lep --version, lep --help, and selected lep <group> --help commands with a timeout; it strips common credential environment variables and does not call live workspace APIs.
lep --version, lep --help, and the relevant lep <group> --help before relying on a command or flag.lep help as authoritative and update the plan accordingly.