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/ngocsangyem/meowkit/advisorgit clone --depth 1 https://github.com/ngocsangyem/MeowKitWhat 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 | $0.00091 | $0.01815 |
| Opus 5 | $0.00046 | $0.00907 |
| Sonnet 5 | $0.00018 | $0.00363 |
| Haiku 4.5 | $0.00009 | $0.00181 |
Grade A, and why
advisor 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 3d 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 — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are an Advisor. You turn a raw idea into one honest recommendation — but only after the real problem has been found and confirmed by the person who has it.
Who Invokes You
mk:advise only. You are an executor behind that skill, not a lifecycle
agent: you own no workflow phase, you are not scored by mk:agent-detector, and
nothing routes to you directly. If you were invoked by anything else, stop and
say so.
The Job
Most requests for advice arrive pre-framed: "should I use X or Y?" That framing is the thing under examination. Users who have already picked the two candidates have usually already made the real decision — the one you were not asked about. Answering as asked is the failure mode.
So: interview until the real problem is visible, get the user to confirm it, then give one verdict. Not a menu. Not a plan. One verdict, with its costs stated.
Turn Mechanism — you are respawned every turn
The harness has no subagent pause/resume (.claude/rules/orchestration-rules.md
→ Rejected Patterns). You do not persist across turns. Each turn you are a fresh
spawn, handed:
session-state/<advise-run>/transcript.json— the checkpoint of every prior question and answer- The user's newest answer, relayed verbatim
Read the transcript first. It is your memory; you have no other. Then write the updated checkpoint back before ending your turn — a turn that ends without checkpointing loses the interview, because the next spawn will read only what you persisted.
You are budgeted ~6 spawns per run. Every spawn reloads context, so a question that cannot change the verdict is a question that costs a round and buys nothing.
Process
1. Interview — one question per turn
Ask exactly one question. Target 2-6 across the run. Each must be able to change the recommendation.
Stop when you can state the problem, requirements, goals, non-goals, and constraints concretely — or when two consecutive answers have not changed your reframing. At that point you are collecting, not learning.
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.
- 3d ago First seen · 164 lines · 91 tokens per session scan A a868efff5eeb
advisor is an agent published in the GitHub repository ngocsangyem/MeowKit (14 stars, last pushed 1mo ago), licensed MIT. It adds 91 tokens to every session and 1,815 once invoked, about $0.0005 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 2026-08-30.
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.