registry / create-lean-agent-kit / 1.0.0

create-lean-agent-kit@1.0.0

Lean core memory kit for AI coding agents — map-first context, convention checks, optional packs for specs/stacks/guardrails.

AI Security Review

scanned 7h ago · by lpm-firewall-ai

LPM treats this as warn-only first-party agent extension lifecycle risk. The CLI copies local templates into a user-selected project when explicitly run. Packaged agent instructions can configure Cursor/Claude project integrations, including optional Cursor session prompts. No exfiltration or remote execution is established.

Static reason

No blocking static signals were detected.

Trigger

Explicit execution of `create-lean-agent-kit` and subsequent explicit `leanagentkit-wire-agent` use

Impact

Can add `.agent` content and, through the agent setup workflow, project-local Cursor/Claude control files.

Mechanism

User-invoked project scaffolding and AI-agent integration setup

Rationale

Source inspection found no install-time execution, network endpoint use, credential collection, remote payload loading, or destructive behavior. Warn because the package intentionally supports project-local AI-agent configuration and session-hook setup.

Evidence

package.jsonbin/cli.mjstemplate/core/.agent/skills/leanagentkit-wire-agent.mdtemplate/core/.agent/install/cursor/hooks.jsontemplate/core/.agent/install/cursor/rules/memory.mdctemplate/core/.agent/install/claude/CLAUDE.md<target>/.agent/**<target>/.leanagentkit-backup/**<target>/.cursor/**<target>/.claude/**

create-lean-agent-kit@1.0.0

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Source & flagged code