Skill of Skills
🔗 mcp serverActiveCuratedcritical

loki-mode

by asklokesh

Multi-agent provider agnostic Autonomous system & framework that WORKS..!

Stars

973

Skill Type

🏗️ Code Scaffolding & Templates

Quality Score

160/200

License

NOASSERTION

Forks

188

Last Updated

Jun 11, 2026

Discovered

Mar 27, 2026

Validation

Passed

github.com/asklokesh/loki-mode

Quality Breakdown

160/ 200

Content Signals

✓Gotchas/Edge Cases+40
✓Progressive Disclosure+30
○Trigger Description+20
✓Verification/Safety+20
✓Code Examples+15
✓Composability+15

Repo Health

✓Recent Activity+15
✓Scripts/Automation+10
○Real Usage (Issues)+10
✓Single Responsibility+10
✓Config/Persistence+10
✓Install Instructions+5

Multi-platform bonus: +5 pts if tool supports 2+ platforms. Score derived from 12 structural signals — not stars or popularity.

Trust & Verification

critical

Manual security review required. Use with extreme caution.

Active

Updated within the last 90 days. Actively maintained.

Unverified skill. Always review source code before installing any skill from an unknown author.

Risk Assessment

  • Autonomous agent system that explicitly operates without human approval: 'You are an autonomous agent. You make decisions. You do not ask questions. You do not stop.'
  • Requires --dangerously-skip-permissions flag, indicating intentional bypass of safety gates
  • Multi-agent spawning capability: '41 specialized types across 8 swarms' with documented agent orchestration
  • Persistent configuration modification: .claude/settings.json, .loki/ directory with state/orchestrator.json, session tracking, and queue management
  • Autonomous execution of bash commands and code deployment: 'Spec in, product out' with automated Git commits and deployment
  • RARV cycle (Reason-Act-Reflect-Verify) enables continuous autonomous decision-making without human-in-loop gates