Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Get Started Free →

Loki Mode - Multi-Agent Autonomous Startup System · dokhacgiakhoa | DecimalAI

agent toolingby dokhacgiakhoa

Loki Mode - Multi-Agent Autonomous Startup System

dokhacgiakhoa/loki-mode

Source dokhacgiakhoa/agent-skills-4-vibe-coding-cli ↗License NOASSERTION

Multi-agent autonomous startup system for Claude Code.

SkillScore95Maintained

Automatic

AI Engineer pack

Last verified 8/29/2026

What it changes

Passed 20 of 21 test cases with the skill, and 15 of 21 without it.

Without skill73%

With skill95%

Checks it passed that the model alone missed

Sonnet model tier is assigned to implementation sub-tasks.
Playwright MCP is specified for automated browser testing.
Opus model tier is assigned strictly to planning sub-tasks.

and 5 more in the full report

5 fixed16 passed either way1 still failing0 broke

Pass-rate lift: +23 pp
Tokens: +58%
Agent turns: 0%
Response time: +16%

Model gemini-3.6-flashMeasured 8/29/2026v121 of 22 cases passed with skill · 21 comparable

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/dokhacgiakhoa-loki-mode/SKILL.md https://app.decimal.ai/s/dokhacgiakhoa-loki-mode/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add dokhacgiakhoa-loki-mode, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/dokhacgiakhoa-loki-mode/ and pins it in .decimal/skills.lock

decimalai skills add dokhacgiakhoa-loki-mode

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/dokhacgiakhoa-loki-mode/SKILL.md https://app.decimal.ai/s/dokhacgiakhoa-loki-mode/SKILL.md

Raw file: /s/dokhacgiakhoa-loki-mode/SKILL.md ↗ · machine-readable summary at /s/dokhacgiakhoa-loki-mode.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 5

Prompt

We need to run comprehensive end-to-end regression tests on our checkout workflow using Playwright. Please route the operational sub-tasks and automated self-critique loops using your agent orchestration framework, then output the execution logs and pass/fail summary.

What was checked, without and with the skill
Checked	Without skill	With skill
Haiku model tier is assigned to execute Playwright test suite operations.	Missed	Passed
Playwright MCP harness pattern is designated for end-to-end browser execution.	Missed	Passed
Automated self-critique review loops are executed before outputting final pass results.	Passed	Passed

case-02 · gemini-3.6-flash · 8/29/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md42 lines18 filesincludes scripts

Loki Mode - Multi-Agent Autonomous Startup System

Version 2.35.0 | PRD to Production | Zero Human Intervention Research-enhanced: OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, Agent SDK, HN Production (2025)

Quick Reference

🧠 Knowledge Modules (Fractal Skills)

1. Critical First Steps (Every Turn)

2. Key Files (Priority Order)

3. Decision Tree: What To Do Next?

4. SDLC Phase Flow

5. Essential Patterns

6. Protected Files (Do Not Edit While Running)

7. Task Tool Model Parameter

8. Opus Task Categories (RESTRICTED - Planning Only)

9. Sonnet Task Categories (Development)

10. Haiku Task Categories (Operations - Use Extensively)

11. Parallelization Strategy

12. Advanced Task Tool Parameters

13. Routing Mode Optimization (AWS Bedrock Pattern)

14. E2E Testing with Playwright MCP (Anthropic Harness Pattern)

15. Efficiency Metrics (Track Every Task)

16. Reward Signals (Learn From Outcomes)

17. Dynamic Agent Selection by Complexity

18. Implementation Anti-Patterns

19. Review Anti-Patterns

20. System Anti-Patterns

21. Always Do These

22. Model-Level Fallbacks

23. Workflow-Level Fallbacks

24. Human Escalation Triggers

25. Loki Mode Constitution

26. Self-Critique Workflow

27. Narrow Scope Wins

28. Confidence-Based Routing

29. Deterministic Outer Loops

Track record

Benchmark

20 / 21

test cases passed, vs 15 / 21 without

Adoption

Users: 1
Times used: 27

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/10/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

dokhacgiakhoa / agent-skills-4-vibe-coding-cli

.agent/skills/loki-mode/SKILL.md

View on GitHub ↗

License NOASSERTIONImported 8/10/2026Version v1Bundled 17 files · scripts

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

loki-modeVersion 2.35.0 | PRD to Production | Zero Human Intervention > Research-enhanced: OpenAI SDK, DeepMind, Anthropic, AWS Bedrock, Agent SDK, HN Production (2025)Close match+36 pp pass-rate lift

More skills from the same source and nearby in the registry, with their headline benchmark lift.

Ready to supercharge your AI agents?

See it in action

What's inside

Loki Mode - Multi-Agent Autonomous Startup System

Quick Reference

🧠 Knowledge Modules (Fractal Skills)

1. Critical First Steps (Every Turn)

2. Key Files (Priority Order)

3. Decision Tree: What To Do Next?

4. SDLC Phase Flow

5. Essential Patterns

6. Protected Files (Do Not Edit While Running)

7. Task Tool Model Parameter

8. Opus Task Categories (RESTRICTED - Planning Only)

9. Sonnet Task Categories (Development)

10. Haiku Task Categories (Operations - Use Extensively)

11. Parallelization Strategy

12. Advanced Task Tool Parameters

13. Routing Mode Optimization (AWS Bedrock Pattern)

14. E2E Testing with Playwright MCP (Anthropic Harness Pattern)

15. Efficiency Metrics (Track Every Task)

16. Reward Signals (Learn From Outcomes)

17. Dynamic Agent Selection by Complexity

18. Implementation Anti-Patterns

19. Review Anti-Patterns

20. System Anti-Patterns

21. Always Do These

22. Model-Level Fallbacks

23. Workflow-Level Fallbacks

24. Human Escalation Triggers

25. Loki Mode Constitution

26. Self-Critique Workflow

27. Narrow Scope Wins

28. Confidence-Based Routing

29. Deterministic Outer Loops

Track record

Benchmark

Adoption

SkillSafety

Source

Other skills for the same job

Related skills