Detonation LabsBehavioral security for the AI agent supply chain

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.

0 findings. Scanned 26 Oct 2026 by detonator 0.5.1.

Target
decision-fatigue-offloader-en
Source
clawhub 1.1.0
SHA-256
df326038aacaf6f4
Files
5
RecommendationNo action needed

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

batch:clawhub:2026-10-26

What this package does

observed across 1 run
  1. Package decision-fatigue-offloader-en 5 files · agent skill
  2. Install No install stepFiles were copied in as they are.
  3. Runs as ls -Rcat references/user-rules.mdcat references/default-rules.mdcat skill-card.md An AI agent also used it 1 time.
  4. What it reached
    Network

    No connections.

    Credentials

    None of the planted fake keys were touched.

    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

No findings. Detonator observed no malicious or suspicious behavior in the runs below.

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.

Show process tree
  • runbash -c cd ./skill && ls -R && echo "-----" && cat references/use...
    • runls -R
    • runcat references/user-rules.md
    • runcat references/default-rules.md
  • runbash -c cd ./skill && cat skill-card.md && echo "-----" && cat _m...
    • runcat skill-card.md
    • runcat _meta.json
  • runbash -c date "+%Y-%m-%d %H:%M %A"
    • rundate +%Y-%m-%d %H:%M %A

Exploration log

1 run
  1. llm:baseline ok · 8.3s

    no new behavior · 4 model steps, 3 commands

    $ cd ./skill && ls -R && echo "-----" && cat references/user-rules.md 2>/dev/null; echo "-----"; cat references/default-rules.md 2>/dev/null $ cd ./skill && cat skill-card.md && echo "-----" && cat _meta.json $ date "+%Y-%m-%d %H:%M %A" model said: I read the skill and its rules files. `user-rules.md` is still empty (no home city, no preferences), so I'm using the default rules and the default city, Beijing. **Decision:** Order a rice-bowl or noodle set for dinner tonight, ¥90 or less per person, delivered or within 1 km, and skip any place with a queue over 10 minutes. **Why:** Beijing's dining multiplier is 1.5, so the ¥60 dinner base be

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.