Decepticon is an autonomous red-team agent that coordinates AI agents, security tools, sandboxes, and supporting services for authorized cybersecurity assessments. Security researchers and red teams can run it through its Docker stack, cloud service, command-line interface, or Python SDK, with the catalogue entries representing its available skills.
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 skills add PurpleAILAB/Decepticon --skill apt41-double-dragongit clone --depth 1 https://github.com/PurpleAILAB/DecepticonWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/purpleailab/decepticon/apt41-double-dragon)<a href="https://agentmods.dev/skills/purpleailab/decepticon/apt41-double-dragon"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/apt41-double-dragon/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/purpleailab/decepticon/apt41-double-dragon"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/apt41-double-dragon.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00053 | $0.05881 |
| Opus 5 | $0.00026 | $0.02941 |
| Sonnet 5 | $0.00011 | $0.01176 |
| Haiku 4.5 | $0.00005 | $0.00588 |
Grade D, and why
apt41-double-dragon scanned grade D with 2 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 10d 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.
Harvests environment variableshighData exfiltration
Enumerating or grepping the environment for keys collects credentials unrelated to what the mod says it does.
- **T1555 / .003** — Harvest credential stores and browser-saved creds (BrowserGhost). Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **T1548.002 / .003** — Bypass UAC on Windows; abuse sudo on Linux. How it starts
The opening of the file, as written. The whole thing — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
APT41 (Double Dragon, Wicked Panda, BARIUM) — Adversary Emulation Profile
APT41 (MITRE ATT&CK G0096; also tracked as Double Dragon, Wicked Panda, Winnti, BARIUM, and Microsoft's Brass Typhoon) is a prolific Chinese threat actor unique for running state-sponsored cyber-espionage in parallel with financially motivated cybercrime — often reusing the same non-public malware for both missions. First publicly profiled by FireEye/Mandiant in August 2019, the group has been active since at least 2012, beginning in the video-game economy (virtual-currency theft, ransomware, code-signing certificate theft) and expanding into wide-ranging espionage. APT41 is best known for software supply-chain compromises (poisoning legitimately signed installers), rapid weaponization of newly disclosed vulnerabilities, abuse of stolen code-signing certificates, and a deep, modular toolset spanning Windows and Linux. This profile maps APT41's documented TTPs to ATT&CK so Decepticon can emulate them under authorized scope and the blue cell can anticipate detection.
Attribution & motivation
- Suspected sponsor / nation: People's Republic of China. Mandiant assesses APT41 operates under the MSS (Ministry of State Security) contractor model — nominally private companies executing intelligence taskings for provincial MSS bureaus — rather than the PLA model. The group is operationally linked to the Chengdu-based front company Chengdu 404 Network Technology Co., Ltd.
- Motivation: Dual-mandate. Espionage (intellectual property, strategic intelligence, surveillance of dissidents/pro-democracy figures) interleaved with financial gain (game virtual-currency manipulation, ransomware, cryptomining). Mandiant observed espionage during Beijing business hours and financially motivated activity during off-hours, suggesting moonlighting on shared infrastructure.
- Confidence: High for China nexus and dual motivation. The U.S. DOJ on September 16, 2020 indicted five Chinese nationals (Zhang Haoran, Tan Dailin, Jiang Lizhi, Qian Chuan, Fu Qiang) and two Malaysian accomplices (Wong Ong Hua, Ling Yang Ching) in connection with attacks on 100+ victims globally, lending strong public attribution. Specific operation-to-operator mapping within individual intrusions carries lower confidence.
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.
- 10d ago First seen · 176 lines · 53 tokens per session scan D c1a0d620430f
apt41-double-dragon is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,471 stars, last pushed 10d ago), licensed Apache-2.0. It adds 53 tokens to every session and 5,881 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it D with 2 findings (harvests environment variables, asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
interactive-dashboard
Interactive web dashboards: stock trackers, sector heatmaps, portfolio monitors — served via preview URL.
onboarding
First-time user onboarding to set up investment profile, watchlists, portfolio, and preferences.
idea-generation
Stock screening and idea generation: quantitative screens, thematic analysis, shortlist.
secretary
Workspace and research management — dispatch analyses, monitor running agents, manage workspaces and threads.
python-lib-analyzer
Analyze any Python library structure, explore modules, classes, and functions with signatures and documentation.
analyzing-windows-prefetch-with-python
Use when parse Windows Prefetch files using the windowsprefetch Python library to reconstruct application execution history, detect renamed or masquerading binaries, and identify suspicious program execution patterns. Use when working with analyzing windows prefetch with python.