Verdict

No malicious behavior observed

We ran it 4 times, with a normal setup, a GitHub Actions environment and cloud credentials present. It made no network connections. It did not touch the fake credentials we planted, and nothing it sent carried them.

0 findings. Scanned 27 Oct 2026 by detonator 0.5.1.

Target: novel-dialogue-subtext-cn
Source: clawhub 1.0.1
SHA-256: c14878061a178754
Files: 14

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 4 runs

Package novel-dialogue-subtext-cn 14 files · agent skill
Install No install stepFiles were copied in as they are.
Runs as python3 <skill>/scripts/self_test.pyls -Rcat SKILL.mdhead -5 An AI agent also used it 1 time.
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 SKILL.md | hea...
- runls -R
- runcat SKILL.md
- runhead -5
- runcat fixtures/success-example.md
- runcat fixtures/failure-missing-input.md
runbash -c cd ./skill && cat README.md && echo "-----" && cat manife...
- runcat README.md
- runcat manifest.yaml
- runsed -n 1,80p scripts/self_test.py
runbash -c cd ./skill && python3 scripts/self_test.py; ls -a .. | he...
- runpython3 scripts/self_test.py
- runhead -30
- runls -a ..
runbash -c cd .. && cat README.md | head -40 && ls -R src | head -40
- runcat README.md
- runhead -40 ×2
- runls -R src
runbash -c pwd; ls -a; cat README.md 2>/dev/null | head -40; ls -R s...
- runls -a
- runcat README.md
- runhead -40 ×2
- runls -R src
runbash -c cat package.json; echo ---; cat src/index.js | head -40; ...
- runcat package.json
- runcat src/index.js
- runhead -40
- rungit log --oneline
- runhead -5

Exploration log

4 runs

scripts:baseline ok · 1.0s

no new behavior
scripts:ci-github ok · 1.2s

no new behavior
scripts:cloud-creds ok · 0.8s

no new behavior
llm:baseline ok · 14.2s

no new behavior · 7 model steps, 6 commands

$ cd ./skill && ls -R && echo "-----" && cat SKILL.md | head -5 && echo "-----" && cat fixtures/success-example.md 2>/dev/null; echo "-----"; cat fixtures/failure-missing-input.md 2>/dev/null $ cd ./skill && cat README.md && echo "-----" && cat manifest.yaml && echo "-----" && sed -n 1,80p scripts/self_test.py $ cd ./skill && python3 scripts/self_test.py; ls -a .. | head -30 $ cd .. && cat README.md | head -40 && ls -R src | head -40 $ pwd; ls -a; cat README.md 2>/dev/null | head -40; ls -R src 2>/dev/null | head -40 $ cat package.json; echo ---; cat src/index.js | head -40; echo ---; git log --oneline | head -5 model said: I read the skill (SKILL.md, both fixtures, and the self-test, which passes). The skill needs a dialogue excerpt plus each character's goal, their relationship, and the fixed facts. Your message only contains the skill's own description, so there's no material to revise yet. ## 未处理项 - 缺少对白片段：没有原句，也没有说话人归属。 - 缺少每位人物的当下目的：想争取、隐瞒、试探还是拒绝。 - 缺少人物关系：亲疏、权力差、旧事或当前冲突。 - 缺少不可改变的事实与语气限制：哪些情节、设定、关系必须原样保留。 -

scripts/scripts/self_test.py: coverage plateau (2 runs with no new behavior); 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.