Command Management Skill
SkillDev toolsUse when a user wants to create, update, list, enable, disable, or delete Space or project slash commands through natural language.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the Command Management Skill skill
What this skill tells your AI
The instructions your AI receives, as published by contextgo/contextgo in src/process/resources/skills/_builtin/command-management/SKILL.md and read by ahel’s review.
You can manage ContextGo commands for either the current project or the bound Space.
Rules
- Ask which scope to use if the user did not make it clear:
projectorspace. - Use
projectfor workspace-private commands and overrides in.contextgo/commands.json. - Use
spacefor shared commands stored with the bound Space. - Update by slash name within the selected scope. If
/reviewalready exists in that scope, overwrite it directly. - Delete by slash name within the selected scope.
- Ask for confirmation in the conversation before destructive deletion or broad overwrites when the user intent is ambiguous.
- Output control commands directly. Do not wrap them in markdown code fences.
Command Formats
List commands
Output:
[COMMAND_LIST: scope=project]
or
[COMMAND_LIST: scope=space]
Create or update a command
Output:
[COMMAND_UPSERT] scope: project name: review enabled: true description: Review the current diff for regressions and missing tests. template: Review the current changes like a strict code reviewer. Prioritize bugs, regressions, risky assumptions, and missing tests. [/COMMAND_UPSERT]
Required fields:
scope:projectorspacename: slash command name without spacesdescription: short user-facing summarytemplate: the prompt template body
Optional fields:
enabled:trueorfalse
Delete a command
Output:
[COMMAND_DELETE: scope=project; name=review]
or
[COMMAND_DELETE: scope=space; name=review]
Guidance
- Commands are user-facing shortcut templates, not autonomous agent skills.
- Keep each command focused on one clear workflow.
- Prefer concise descriptions and explicit templates.
- If the user gives only intent, infer a clean
descriptionandtemplatebefore writing the command.
Signals
- GitHub stars
- 54
- Forks
- 5
- Last commit
- May 2026
Advanced
- Catalog kind
- skill
- Gateway key
command-management- Source
- github.com/contextgo/contextgo