Skip to content
Security
Skill

/plamen

Launch the Plamen deterministic Web3 security audit pipeline

From plugin
plamen
276160 skills12 agents4 commands
Install
$ npx -y skills add PlamenTSV/plamen --skill plamen --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/plamen

Context preview

The summary Claude sees to decide when to auto-load this skill.

Launch the Plamen deterministic Web3 security audit pipeline

SKILL.md

plamen.SKILL.md
name: plamen
description: "Launch the Plamen deterministic Web3 security audit pipeline"

Plamen V2 Wizard Launcher For Codex

Use this skill whenever the user invokes `$plamen`, `/plamen`, asks to start, resume, or configure a Plamen audit, or asks for the Plamen wizard inside Codex.

Hard Rule

Do not manually orchestrate Plamen phases. Do not spawn recon, breadth, depth, verification, or report agents yourself. The Python driver is the sole owner of phase sequencing for both Claude and Codex routes.

Your job is the same job as the Claude `/plamen` command wizard:

1. Detect an existing audit and offer resume/fresh/new. 2. Collect missing launch parameters. 3. Write or reuse `{PROJECT_ROOT}/.scratchpad/config.json`. 4. Launch the deterministic driver. 5. Report the resume command and basic status.

For new Codex launches, `config.json` must set `"cli_backend": "codex"`. For existing audits, do not rewrite the config on resume.

Wizard Files

Follow the Codex-native wizard references in this skill directory:

  • Smart-contract audits: `plamen-wizard.md`
  • L1 infrastructure audits: `plamen-l1-wizard.md`

Read only the relevant file:

  • If the user says `l1`, `L1`, `infra`, `client`, `go`, `rust node`, or

the target looks like a chain/client codebase, use `plamen-l1-wizard.md`.

  • Otherwise use `plamen-wizard.md`.

Invocation Syntax

$plamen [l1] [light|core|thorough] [path] [docs:<path-or-url>] [scope:<path>] [notes:<text>] [--fresh]
$plamen resume [path-or-config]

Defaults:

  • `pipeline`: `sc`
  • `mode`: `core`
  • `project_root`: current working directory
  • `cli_backend`: `codex`

Do not ask a model-selection question from this skill. The user is already running inside the model/backend they chose.

Driver Commands

Codex route:

python C:\\Users\\plmnt\\.codex\\plamen\\scripts\plamen_driver.py "{CONFIG_PATH}"

Fresh restart:

python C:\\Users\\plmnt\\.codex\\plamen\\scripts\plamen_driver.py --fresh "{CONFIG_PATH}"
Read more
Ships withplamen

Autonomous Web3 security auditor for Claude Code and OpenAI Codex CLI. Orchestrates 18-100 AI agents across 40+ phases to produce audit reports with verified PoC exploits — for smart contracts and L1 node-client infrastructure.

Get the whole plugin