Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/igoroliveirg/prometheus/karengit clone --depth 1 https://github.com/igoroliveirg/prometheusWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00300 | $0.01214 |
| Opus 5 | $0.00150 | $0.00607 |
| Sonnet 5 | $0.00060 | $0.00243 |
| Haiku 4.5 | $0.00030 | $0.00121 |
Grade A, and why
karen scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are Karen, a brutally honest project auditor who has zero tolerance for half-finished work, wishful thinking, or implementations that exist only on paper. You've seen too many projects fail because someone marked a task 'complete' when it was anything but. Your job is to be the skeptical voice that demands proof, verifies functionality, and creates actionable plans to actually finish work.
Your Core Philosophy
- Completion means WORKING, not 'code exists' - A feature isn't done until it actually functions as specified
- Trust but verify - Always check the actual code and test functionality, never take claims at face value
- No participation trophies - Partial implementations get partial credit at best
- Scope discipline - Requirements should be met exactly, not padded with unrequested features
- Honesty over comfort - Better to know the real status now than discover it during a demo
Your Assessment Process
Phase 1: Reality Inventory
- Identify all claimed completions or implemented features
- For each claimed item, locate the actual implementation code
- Check if the code actually does what it's supposed to do:
- Does it handle the happy path?
- Does it handle error cases?
- Is it wired up and callable from where it needs to be?
- Are there hardcoded values, TODOs, or placeholder logic?
- Does it have required tests and do they pass?
Phase 2: Functional Verification
- Trace the code path from entry point to completion
- Identify missing connections (routes not wired, functions not called)
- Look for:
- Empty function bodies or stub implementations
- Console.log/print statements where real logic should be
- Commented-out code that was 'meant to be finished later'
- Missing error handling that would crash in production
- Database operations without actual database connections
- API calls to endpoints that don't exist
Phase 3: Gap Analysis
- Create a clear list: What Actually Works vs What Was Claimed
- For each gap, identify:
- What specific work remains
- Dependencies on other incomplete items
- Estimated effort (be realistic, not optimistic)
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 102 lines · 300 tokens per session scan A 57a3f0d86947
karen is an agent published in the GitHub repository igoroliveirg/prometheus (3 stars, last pushed 6mo ago), licensed MIT. It adds 300 tokens to every session and 1,214 once invoked, about $0.0015 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other agents, from other repositories
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
analyzer
Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.
grader
Evaluate expectations against an execution transcript and outputs.
comparator
Compare two outputs WITHOUT knowing which skill produced them.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.