agent-kernel-ops
Skill5 starsChanged 4 months agoMITView on GitHub
Operational guide for agents working with Agent Kernel v1.20.0+ — environment vault, memory proposals, policies, installation, and diagnostics
- Reads credentials
- Installs packages
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- .codex/AGENTS.mdAGENTS.md
- agent-kernelSkill
- architecture-guardianSkill
- agent-kernelSkill
- architecture-guardianSkill
- agent-kernelSkill
- agent-kernel-evolveSkill
- architecture-guardianSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
16 rules. About 0 are common, 9 are distinctive, and 7 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈9
Distinctive · estimate
This author's own choices. Worth a look.
7
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Use this skill when the user mentions agent-kernel, .env, Environment Vault, memory proposals, agent rules, policies, or runtime diagnostics
- Distinctive: line 12
Never print, summarize, diff, log, or paste environment file contents
- Very specific: line 14
Use agent-kernel env doctor [project] when the vault is unhealthy or permissions drift
- Distinctive: line 15
Use normal pull for missing files
- Very specific: line 16
Do not use --force until the user understands that a differing local file will be replaced
- Distinctive: line 17
Keep backups enabled during forced restore unless the user explicitly requests otherwise
- Very specific: line 18
Use unlink to detach a project path while retaining stored files
- Very specific: line 19
Use purge --yes only after an explicit request to delete the stored vault
- Distinctive: line 38
Check identity, health, and file states
- Distinctive: line 70
Run health checks and repair POSIX permissions
- Distinctive: line 113
Delete stored data only after explicit confirmation
- Very specific: line 125
Read the complete command and security guide at docs/ENVIRONMENT_VAULT.md
- Very specific: line 145
If agent-kernel is not found after Bun install, add Bun's bin directory to PATH:
- Distinctive: line 182
Run file and policy guards
- Distinctive: line 188
Verify a named policy
- Distinctive: line 202
Run the main health doctor
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Agent Kernel Universal Operations Guide
Use this skill when the user mentions agent-kernel, .env, Environment Vault, memory proposals, agent rules, policies, or runtime diagnostics
Environment Vault safety rules
- Never print, summarize, diff, log, or paste environment file contents
- Start with
agent-kernel env status [project] - Use
agent-kernel env doctor [project]when the vault is unhealthy or permissions drift - Use normal pull for missing files
- Do not use
--forceuntil the user understands that a differing local file will be replaced - Keep backups enabled during forced restore unless the user explicitly requests otherwise
- Use
unlinkto detach a project path while retaining stored files - Use
purge --yesonly after an explicit request to delete the stored vault - Reject symlinks, non-regular files, paths outside the project root, and unstable project identity
Environment Vault commands
Link the project and discover eligible environment files
agent-kernel env link [project]Link exact Monorepo files
agent-kernel env link [project] \
--include apps/api/.env \
--include apps/web/.env.localCheck identity, health, and file states
agent-kernel env status [project]
agent-kernel env status [project] --jsonStore local edits
agent-kernel env push [project]
agent-kernel env push [project] --file apps/api/.envPreview without writing
agent-kernel env push [project] --dry-runRestore missing files
agent-kernel env pull [project]Handle an intentional overwrite with a backup
agent-kernel env pull [project] --forceRun health checks and repair POSIX permissions
agent-kernel env doctor [project]
agent-kernel env doctor [project] --repair-permissionsMigrate a matching legacy vault after inspection
agent-kernel env doctor [project] --migrateInspect vault health in JSON
agent-kernel env doctor [project] --jsonWatch environment files edited outside an Agent Kernel hook
agent-kernel env watch [project]Inspect revisions without secret contents
agent-kernel env history [project] --file .envRestore one revision with normal conflict handling
agent-kernel env restore [project] --file .env --revision <revision-id>Detach automatic behavior while retaining stored data
agent-kernel env unlink [project]Delete stored data only after explicit confirmation
agent-kernel env purge [project] --yesList all local vault identities
agent-kernel env list --jsonRead the complete command and security guide at docs/ENVIRONMENT_VAULT.md
Installation
With npm (requires Node.js >=18.18.0):
npm install -g @mamdouh-aboammar/agent-kernel
agent-kernel --version
agent-kernel init --syncWith Bun:
bun install -g @mamdouh-aboammar/agent-kernel
agent-kernel --version
agent-kernel init --syncIf agent-kernel is not found after Bun install, add Bun's bin directory to PATH:
export PATH="$HOME/.bun/bin:$PATH"One-off (no global install):
npx -y @mamdouh-aboammar/agent-kernel --version
bunx @mamdouh-aboammar/agent-kernel --versionCurrent stable release: 1.20.0
Memory and rule proposals
Propose a new rule
agent-kernel propose --from <agentName> --text "Your rule text" --reason "Why"Inspect the approval inbox
agent-kernel inboxApprove and publish a proposal
agent-kernel approve <proposalId> --publishGovernance and quality guards
Run file and policy guards
agent-kernel guard [--staged|--file path]Verify a named policy
agent-kernel policy check mandatory-bun-package-managerHealth and context diagnostics
Check Kernel status
agent-kernel statusRun the main health doctor
agent-kernel doctorDoes it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

