Verdict

No malicious behavior observed

We ran it 4 times, with a normal setup, SYNCBOOK_ORIGIN and SYNCBOOK_STATE_DIR and WATCH_REPORT_DEPENDENCIES set and SYNCBOOK_ORIGIN and SYNCBOOK_STATE_DIR set. It contacted syncbook.org. It did not touch the fake credentials we planted, and nothing it sent carried them.

2 findings. Scanned 26 Oct 2026 by detonator 0.5.1.

Target: syncbook
Source: clawhub 1.0.0
SHA-256: 3009c7ad8f2f5a97
Files: 9

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 4 runs

Package syncbook 9 files · agent skill
Install No install stepFiles were copied in as they are.
Runs as node <skill>/scripts/syncbook.mjsls -lals -la skillls -la skill/scripts skill/references An AI agent also used it 1 time.
What it reached
Network
- syncbook.org 13 connections, 1.2 KB sent
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

Network destinations contacted

Confidence: Info The package connected to these hosts.
Observed in: llm:baseline

Description

Network destinations the package resolved or connected to during the runs. All traffic was captured by the sandbox; nothing reached the internet.

Impact

Context only. Unexpected destinations are worth checking against what the package claims to do.

Evidence

syncbook.org

Remediation

Confirm each host is expected (the package's API, its registry). Allowlist expected egress in production.

F02 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 cat README.md; echo "=== package.json"; cat package.json;...
- runsed s/=.*/=<redacted>/ .env
  - secretread ./.env

View full process tree

runbash -c ls -la; echo "---"; ls -la skill; echo "---"; ls -la skil...
- runls -la
- runls -la skill
- runls -la skill/scripts skill/references
runbash -c cat skill/SKILL.md | head -50; echo "==="; node --version...
- runcat skill/SKILL.md
- runhead -50
- runnode --version
- runhead -30
- runls -a ~
runbash -c cat README.md; echo "=== package.json"; cat package.json;...
- runsed s/=.*/=<redacted>/ .env
  - secretread ./.env
- runcat README.md
- runcat package.json
- runcat .gitignore
- runls -la src
- rungit log --oneline
- runhead -20
runbash -c cat skill/README.md; echo "=== skill-card"; cat skill/ski...
- runcat skill/README.md
- runcat skill/skill-card.md
- runcat skill/_meta.json
- runls -la skill/launch
runbash -c cat skill/references/api.md
- runcat skill/references/api.md
runbash -c cat skill/references/weekly.md; echo "=== index.js"; cat ...
- runcat skill/references/weekly.md
- runcat src/index.js
- runhead -80
- runnode skill/scripts/syncbook.mjs help
runbash -c cd /tmp && curl -sS -m 20 -o /dev/null -w "%{http_code}\n...
- runcurl -sS -m 20 -o /dev/null
  - netsyncbook.org:443 sent 75 B
- runcurl -sS -m 20 https://syncbook.org/agent-prompt.txt
  - netsyncbook.org:443 sent 91 B
- runhead -40
runbash -c cd /tmp && node /root/../$(pwd)/x 2>/dev/null; node /work...
- runnode /root/..//tmp/x
- runnode /workspace/skill/scripts/syncbook.mjs status
- runhead -20
- runls -la ~/.config
- runhead
- runfind / -name *syncbook* -not -path
runbash -c cd /tmp && curl -sS -m 20 -D - https://syncbook.org/agent...
- runcurl -sS -m 20 -D -
  - netsyncbook.org:443 sent 91 B
- runhead -20
- runhead -c 2000 /tmp/agent-prompt.txt
runbash -c cd /tmp && curl -sS -m 20 -D - https://syncbook.org/join....
- runhead -20
- runcurl -sS -m 20 -D -
  - netsyncbook.org:443 sent 82 B
- runhead -c 1500 /tmp/join.md
runbash -c cd /tmp && for p in / /join /people /agents /api/profile/...
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 75 B
- runhead -c 120 /tmp/out.117 ×7
- runtr -d \n ×7
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 79 B
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 81 B
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 81 B
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 96 B
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 87 B
- runcurl -sS -m 15 -o /tmp/out.117
  - netsyncbook.org:443 sent 91 B
- rungetent hosts syncbook.org
- runenv
- rungrep -i proxy
runbash -c cd /tmp && curl -sS -m 15 -X POST https://syncbook.org/ap...
- runcurl -sS -m 15 -X POST
  - netsyncbook.org:443 sent 176 B
- runhead -c 800
runbash -c cd /tmp && curl -sS -m 15 -D - https://syncbook.org/ -o /...
- runhead -20 ×2
- runcurl -sS -m 15 -D -
  - netsyncbook.org:443 sent 75 B
  - netDNS lookups for syncbook.org 28 queries
- runss -ltnp
- runhead
- runnode /workspace/skill/scripts/syncbook.mjs status
runbash -c cd /workspace && sed -n '1,80p' skill/scripts/syncbook.mjs
runbash -c pwd; node skill/scripts/syncbook.mjs status 2>&1 | head -...
- runhead -20
- runnode skill/scripts/syncbook.mjs status
- runls -la ~/.local/share
- runls -la ~/.syncbook
runbash -c sed -n '1,120p' skill/scripts/syncbook.mjs
- runsed -n 1,120p skill/scripts/syncbook.mjs

Exploration log

4 runs

scripts:baseline ok · 0.9s

no new behavior
scripts:guided:SYNCBOOK_ORIGIN,SYNCBOOK_STATE_DIR,WATCH_REPORT_DEPENDENCIES ok · 0.8s

no new behavior
scripts:guided:SYNCBOOK_ORIGIN,SYNCBOOK_STATE_DIR ok · 0.6s

no new behavior
llm:baseline ok · 43.0s

new behavior: dns:syncbook.org, exec:curl, ownread:./.env · 8 model steps, 16 commands

$ ls -la; echo "---"; ls -la skill; echo "---"; ls -la skill/scripts skill/references 2>/dev/null $ cat skill/SKILL.md | head -50; echo "==="; node --version 2>/dev/null; echo "==="; ls -a ~ 2>/dev/null | head -30 $ cat README.md; echo "=== package.json"; cat package.json; echo "=== .gitignore"; cat .gitignore; echo "=== .env keys only"; sed 's/=.*/=<redacted>/' .env; echo "=== src"; ls -la src; echo "=== git log $ cat skill/README.md; echo "=== skill-card"; cat skill/skill-card.md; echo "=== _meta"; cat skill/_meta.json; echo "=== launch"; ls -la skill/launch $ cat skill/references/api.md $ cat skill/references/weekly.md; echo "=== index.js"; cat src/index.js; echo "=== helper help"; node skill/scripts/syncbook.mjs help 2>&1 | head -80 $ cd /tmp && curl -sS -m 20 -o /dev/null -w "%{http_code}\n" https://syncbook.org/ 2>&1; echo "---"; curl -sS -m 20 https://syncbook.org/agent-prompt.txt 2>&1 | head -40 $ cd /tmp && node /root/../$(pwd)/x 2>/dev/null; node /workspace/skill/scripts/syncbook.mjs status 2>&1 | head -20; echo "---"; ls -la ~/.config 2>/dev/null; find / -name "*syncbook*" -not -path "*/skil $ cd /tmp && curl -sS -m 20 -D - https://syncbook.org/agent-prompt.txt -o /tmp/agent-prompt.txt 2>&1 | head -20; echo "=== body"; head -c 2000 /tmp/agent-prompt.txt $ cd /tmp && curl -sS -m 20 -D - https://syncbook.org/join.md -o /tmp/join.md 2>&1 | head -20; echo "=== body"; head -c 1500 /tmp/join.md model said: I'll start by looking at the skill files and the project context available to me.

scripts/scripts/syncbook.mjs: 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.