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 bounty-reportgit 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/bounty-report)<a href="https://agentmods.dev/skills/purpleailab/decepticon/bounty-report"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/bounty-report/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/bounty-report"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/bounty-report.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00046 | $0.01277 |
| Opus 5 | $0.00023 | $0.00639 |
| Sonnet 5 | $0.00009 | $0.00255 |
| Haiku 4.5 | $0.00005 | $0.00128 |
Grade A, and why
bounty-report-formatter 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 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.
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 — 147 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bug Bounty Report Formatter
Format validated findings as platform-ready bug bounty reports that survive triage. Optimized for acceptance rate, not word count. The standard this targets: 7.00 HackerOne signal — precision over volume.
Title Convention
Format: [Component]: [Vulnerability Class] via [Mechanism]
Good titles (from real accepted advisories):
zrok: WebDAV drive backend follows symlinks outside DriveRoot, enabling host filesystem read/writeParse Server: cloud function validator bypass via prototype chain traversalDirectus: TUS Upload Authorization Bypass Allows Arbitrary File Overwritelodash: Code Injection via _.template imports key namesAVideo: Unauthenticated SSRF via HTTP redirect bypass in LiveLinks proxyOpenFGA: Unauthenticated playground endpoint discloses preshared API key
Rules:
- Component name first (the project/package name)
- Vulnerability class in plain English (not CWE numbers in the title)
- Mechanism must be specific (not "via user input" — say "via prototype chain traversal")
- No adjectives like "critical" or "severe" in the title
- Under 120 characters
Report Template
Write to workspace/findings/BOUNTY-{finding_id}.md:
# {Title}
## Summary
{One paragraph. State what the bug is, where it lives (file:line or endpoint),
and what impact it has. Three sentences maximum.}
## Severity
**CVSS 3.1**: {full vector string} ({score} {severity_label})
| Metric | Value | Justification |
|--------|-------|---------------|
| AV | Network | Exploitable over HTTP |
| AC | Low | No special conditions |
| PR | None | No authentication required |
| UI | None | No user interaction |
| S | Unchanged | Impact limited to vulnerable component |
| C | High | Full read access to filesystem |
| I | None | No write capability demonstrated |
| A | None | No availability impact |
## Affected Version
- Package: {name}
- Version: {exact version or range}
- Commit: {commit hash if applicable}
## Steps to Reproduce
1. {Exact setup step — e.g., "Clone the repository: `git clone ...`"}
2. {Exact action — e.g., "Start the server: `npm start`"}
3. {Exact exploit — e.g., "Send the following request:"}
## Proof of Concept
\```bash
{Exact command that demonstrates the vulnerability}
\```
**Expected response (vulnerable):**
\```
{Exact output showing exploitation}
\```
**Baseline response (not vulnerable):**
\```
{Output from the same request without the payload}
\```
## Impact
{What an attacker can actually do with this bug. Only state impacts you
demonstrated in the PoC. Do NOT extrapolate to theoretical scenarios.}
## Remediation
{Specific code fix. Show the diff or the corrected code.}
\```diff
- const user = db.query(`SELECT * FROM users WHERE id=${req.params.id}`)
+ const user = db.query(`SELECT * FROM users WHERE id=$1`, [req.params.id])
\```
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 · 147 lines · 46 tokens per session scan A 307a9ea4582d
bounty-report-formatter is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,471 stars, last pushed 10d ago), licensed Apache-2.0. It adds 46 tokens to every session and 1,277 once invoked, about $0.0002 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 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.
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.
implementing-zero-trust-with-beyondcorp
Use when deploy Google BeyondCorp Enterprise zero trust access controls using Identity-Aware Proxy (IAP), context-aware access policies, device trust validation, and Access Context Manager to enforce identity and posture-based access to GCP resources and internal applications. Use when deploying google beyondcorp…