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 unbounded-consumptiongit 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/unbounded-consumption)<a href="https://agentmods.dev/skills/purpleailab/decepticon/unbounded-consumption"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/unbounded-consumption/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/unbounded-consumption"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/unbounded-consumption.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 4 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Tool Misuse · line 37 Tool calls are chained to bypass individual safety checks or escalate capabilities beyond what any single tool call would allow.Fix: Limit tool chaining depth and validate the output of each tool before passing it to the next. Require explicit user approval for multi-step chains.
- medium Data Exfiltration · line 37 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 107 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
- medium Data Exfiltration · line 111 Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.Fix: Verify the destination URL is trusted and necessary. Remove or replace with documented APIs. Ensure no secrets, tokens, or PII are transmitted.
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 | $0.00054 | $0.01722 |
| Opus 5 | $0.00027 | $0.00861 |
| Sonnet 5 | $0.00011 | $0.00344 |
| Haiku 4.5 | $0.00005 | $0.00172 |
Grade A, and why
unbounded-consumption scanned grade A with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Sends data to an external URLlowData exfiltration
A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.
TOK=$(curl -X POST $TARGET/signup -d '{"email":"test+'$(uuidgen)'@example"}' | jq -r .token) Downgraded: this mod is about security review, or the phrase is quoted, so it is likely naming the pattern rather than instructing it.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
seq 1 1000 | xargs -I{} curl -s -X POST "$TARGET/chat" \ How it starts
The opening of the file, as written. The whole thing — 171 lines — stays where its author put it; the contents beside it link to each section on GitHub.
LLM Unbounded Consumption (LLM10:2025)
LLM inference is metered in dollars-per-token at the provider, and those tokens stack quickly: a context window full of attacker content costs more than the rest of the request stack combined. Unbounded consumption produces three impacts in escalating severity: provider rate-limit / hard-block (DoS), bill blowout (denial-of- wallet), and ultimately tool / sandbox resource exhaustion (DoS of the customer's compute).
1. Recognition signals
- The product exposes an authenticated or unauthenticated LLM endpoint that accepts large prompts.
- Per-user / per-tenant token budget is undocumented or absent.
- Free-tier signup grants immediate access to the most expensive model.
- Agentic system has no max-step / max-token / max-cost cap.
- Tools loop on model output without iteration cap (
while not done:). - File-upload feature dumps full document into the context.
- Background workers retry failed model calls on exponential backoff without a hard ceiling.
- Cost dashboard updates daily, not in real time.
2. Attack vectors
Direct prompt expansion (input DoS)
Submit a maximum-context-window prompt repeatedly:
seq 1 1000 | xargs -I{} curl -s -X POST "$TARGET/chat" \
-H "Authorization: Bearer $FREE_TIER_TOKEN" \
-d "{\"prompt\":\"$(python -c 'print("repeat this " * 30000)')\"}" \
>/dev/null &
Cost-tier escalation
Bypass the model picker to force the most expensive model (opus / o1 / claude-3.7) on every request. Often the picker is a client-side selector that the backend trusts.
Runaway agentic loop
Submit a task that the agent cannot complete: "Read every file in
/ recursively and summarise each in 5 paragraphs." Each tool
result feeds the next prompt; tokens grow per loop. With no max-step
cap the run lasts until provider rate-limits or budget alarms fire.
Fan-out via tool calls
Trigger an LLM that itself spawns N tool calls per turn, each of which invokes a sub-LLM. Geometric blow-up.
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.
- 9d ago First seen · 171 lines · 54 tokens per session scan A 30a8bf326c2c
unbounded-consumption is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,491 stars, last pushed 13d ago), licensed Apache-2.0. It adds 54 tokens to every session and 1,722 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 2 findings (sends data to an external url, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
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.