icloudza/algokiller-plugin/general

Claude Code and agent skill by icloudza/algokiller-plugin

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Strong activation entry for AlgoKiller general trace-analysis mode. Bind an ARM64 trace, force-load the trace-analysis methodology, and answer field semantics / execution flow / detection-point / data-flow questions.

Install

npx vskill@latest install icloudza/algokiller-plugin/general

SKILL.md

You are now entering AlgoKiller general trace-analysis mode. Open-ended questions over ARM64 trace evidence. Discipline still matters — drift = hallucination.

MANDATORY FIRST STEPS — execute in order:

  1. Load the full methodology from the ak:trace-analysis skill. Read it completely before any tool call. It contains search-key selection rules, single-purpose-per-search discipline, call-boundary parsing rules, field-semantics layering, execution-flow extraction, and detection-point analysis methodology.
  1. Parse the user input below. The first whitespace-separated token is the absolute path to the ARM64 trace log; the rest is the task description.
  1. Call MCP tool ak.bind_trace with that path and mode="general". Do not proceed if bind fails.
  1. Proceed with the methodology from the skill. Every trace_search / trace_context return will include a discipline_reminder field — read it and respect it.

NON-NEGOTIABLE RULES (skill expands each):

  • Pick a SINGLE purpose per search (locate / origin / consumer / branch / boundary / verify / exclude). Do not reuse one result for multiple roles.
  • Open call / hexdump / ret hits as data-flow boundaries: function name, args, return, hexdump address/length/bytes, x0–x7 set-before, consumers-after.
  • hexdump ASCII is a search hint only — field boundaries come from left-side hex bytes + address + length.
  • Separate confirmed / high-confidence inference / open question in every finding.
  • Do NOT ask the user for missing field names / business semantics / extra samples — search and infer first.
  • Stop once the user's actual question is answerable; one cross-check pass on the key conclusion is enough.
  • write_artifact with .py source ONLY if the task requires code reproduction. Otherwise deliver text/markdown directly, or write .md report for long deliverables.

User input:

$ARGUMENTS

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
59
Repository forks
18
7-day trend score
20

Stars and forks are for the whole GitHub repository, not this skill alone. Refreshed 29 May 2026, 5 months ago.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/icloudza/algokiller-plugin/general/badge)](https://verified-skill.com/skills/icloudza/algokiller-plugin/general)

Something wrong with this skill or its page? Tell us.