Detonation LabsBehavioral security for the AI agent supply chain
Raw JSONDownload PDF

Verdict

No malicious behavior observed

We ran it 1 time, with a normal setup. It made no network connections. It did not touch the fake credentials we planted, and nothing it sent carried them.

1 finding. Scanned 27 Oct 2026 by detonator 0.5.1.

Target
mobile-security
Source
clawhub 1.0.1
SHA-256
cad176f2a33444b1
Files
7
RecommendationNo action needed

Nothing observed calls for action. This is evidence, not a guarantee.

batch:clawhub:2026-10-27

What this package does

observed across 1 run
  1. Package mobile-security 7 files · agent skill
  2. Install No install stepFiles were copied in as they are.
  3. Runs as ls -lafind . -maxdepth 2 -not -pathhead -100ls -la ./skill An AI agent also used it 1 time.
  4. What it reached
    Network

    No connections.

    Credentials
    • ./.envread
    Files written

    Nothing outside its own folder.

We planted fake AWS, SSH, npm and GitHub keys and a .env file before every run. Red means something left or was planted to run later.

Findings

F01 Low

Loaded the project's .env file

Confidence
Info The package loaded the project's .env file.
Observed in
llm:baseline

Description

The process loaded the project's .env file. Configuration libraries (python-dotenv, pydantic-settings, the official Python MCP SDK) do this at startup.

Impact

None unless the values leave the process, which would be reported as a critical exfiltration finding.

Evidence

./.env (by sed)

Remediation

No action needed. Keep long-lived secrets out of project .env files where possible.

Process tree

run llm:baseline

Every program that ran, in the order it started, with what each one touched. Red marks the path to a leak or a planted file.

  • runbash -c cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b...
    • runsed s/=.*/=<redacted>/ .env
      • secretread ./.env
View full process tree
  • runbash -c pwd && ls -la && echo "---" && find . -maxdepth 2 -not -p...
    • runls -la
    • runfind . -maxdepth 2 -not -path
    • runhead -100
  • runbash -c ls -la ./skill 2>/dev/null && echo "---REFS---" && ls -la...
    • runls -la ./skill
    • runls -la ./skill/references
  • runbash -c cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b...
    • runsed s/=.*/=<redacted>/ .env
      • secretread ./.env
    • runcat README.md
    • runcat package.json
    • runcat .gitignore
  • runbash -c cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b...
    • runcat -n src/index.js
  • runbash -c cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b...
    • rungit ls-files
    • rungit branch -a
    • rungit log --oneline -20
    • runfind . -not -path ./.git/* -not
  • runbash -c cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b...
    • runfind . -not -path ./.git/* (
    • runsort -u
    • rungrep -Ei \.(kt|java|swift|dart|plist|gradle|xml)$|pubspec|Podfile|...
    • rungit log --all --name-only --pretty=format:
  • runbash -c cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b...
    • rungit ls-files --error-unmatch .env
    • rungit log --all --oneline -- .env
    • rungrep .
    • runhead -20
    • rungit rev-list --all
    • rungit grep -nEi sk_live|sk_test|api[_-]?key|secret|password|token|BEGIN .... b747999d77f116f3c1e648cc75e6cd5631a4b722 a4e97538c47c87e4686a01dd

Exploration log

1 run
  1. llm:baseline ok · 22.7s

    new behavior: ownread:./.env · 5 model steps, 7 commands

    $ pwd && ls -la && echo "---" && find . -maxdepth 2 -not -path './.git/*' -not -path './skill/*' | head -100 $ ls -la ./skill 2>/dev/null && echo "---REFS---" && ls -la ./skill/references 2>/dev/null $ cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b0c/work && echo "=== README ===" && cat README.md && echo && echo "=== package.json ===" && cat package.json && echo && echo "=== .gitignore == $ cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b0c/work && echo "=== src/index.js ===" && cat -n src/index.js $ cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b0c/work && echo "=== all tracked files ===" && git ls-files && echo && echo "=== branches ===" && git branch -a && echo && echo "=== recent com $ cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b0c/work && echo "=== mobile artifact search ===" && find . -not -path './.git/*' \( -name '*.kt' -o -name '*.java' -o -name '*.swift' -o -name $ cd /var/tmp/detonator-runs/6f55ea86ca6d40e8a88721f28c433b0c/work && echo "=== is .env tracked? ===" && (git ls-files --error-unmatch .env 2>/dev/null && echo "TRACKED" || echo "untracked (good)") && e model said: ## Security audit: payments-api **Result: no mobile application exists in this repository, so there is nothing to audit against OWASP MASVS v2 / Mobile Top 10 2024.** I'm reporting that rather than inventing findings. Scope: entire repo at `b747999` (branch `main`, 2 commits). Files: `src/index.js`, `package.json`, `README.md`, `.gitignore`, `.env`. Platforms: **none** — this is a Node.js/Expres

llm: variants exhausted

Note

Observed behavior only. A clean result does not prove a package is safe. Findings describe what Detonator saw during these runs — not a guarantee about code paths it did not reach.