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

Verdict

No malicious behavior observed

We ran it 4 times, with a normal setup, the environment variables it says it needs and a GitHub Actions environment. It contacted api.juicer.io and developers.juicer.io. It read some credential files, but nothing it sent carried them.

3 findings. Scanned 26 Oct 2026 by detonator 0.5.1.

Target
juicer-social-listening
Source
clawhub 1.0.0
SHA-256
00731880a8292a8e
Files
2
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
  1. Package juicer-social-listening 2 files · agent skill
  2. Install No install stepFiles were copied in as they are.
  3. Runs as ls -lals -la skillhead -50cat skill/SKILL.md An AI agent also used it 1 time.
  4. What it reached
    Network
    • api.juicer.io 6 connections, 583 B sent
    • developers.juicer.io 4 connections, 470 B sent
    Credentials
    • ./.envread
    • ~/.config/gh/hosts.ymlread
    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

api.juicer.io
developers.juicer.io

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 cat)

Remediation

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

F03 Low

Agent read credential files (may be the agent's own choice)

Confidence
Info A credential file was opened by a program that does not own it.
Observed in
llm:baseline

Description

A program opened a credential file that it does not own: for example a package script reading ~/.npmrc, ~/.aws/credentials or ~/.ssh/id_rsa, rather than npm, the AWS CLI or ssh reading their own files. The reading program is named in the evidence. In this run the contents were not seen leaving the sandbox.

Impact

Reading credentials is the collection step that precedes theft. Exfiltration may be gated on conditions this run did not meet (a different date, CI variables, a reachable network), or the data may be staged for a later step.

Evidence

~/.config/gh/hosts.yml (by cat)

Remediation

Review why this package needs the file. Legitimate tools read only their own configuration; anything else should be treated as hostile until explained. Prefer running it with credentials unavailable (separate user, container, or scoped environment).

References

  • ATT&CK T1552.001 Credentials in Files
  • CWE-200 Exposure of Sensitive Information
  • OWASP MCP01:2025 Token Mismanagement & Secret Exposure

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 .env; echo "---"; cat README.md; echo "---"; cat pack...
    • runcat .env
      • secretread ./.env
  • runbash -c cat src/index.js; echo "---"; git log --oneline | head -5...
    • runcat ~/.config/gh/hosts.yml
      • secretread ~/.config/gh/hosts.yml
View full process tree
  • runbash -c ls -la; echo "---"; ls -la skill 2>/dev/null; echo "---";...
    • runls -la
    • runls -la skill
    • runhead -50
    • runcat skill/SKILL.md
  • runbash -c env | grep -i -E "juicer|mcp" ; echo "---"; ls -la ~/.ope...
    • runenv
    • rungrep -i -E juicer|mcp
    • runls -la ~/.openclaw
    • runhead -50
    • runcat ~/.openclaw/*.json
  • runbash -c cat .env; echo "---"; cat README.md; echo "---"; cat pack...
    • runcat .env
      • secretread ./.env
    • runcat README.md
    • runcat package.json
    • runls -la src
    • runcat skill/_meta.json
  • runbash -c which juicer npx node python3 curl 2>/dev/null; echo "---...
    • runwhich juicer npx node python3 curl
    • runls ~/.config
    • runhead -20
    • runfind / -name *juicer* -maxdepth 6
    • rungrep -v proc
  • runbash -c cat src/index.js; echo "---"; git log --oneline | head -5...
    • runcat ~/.config/gh/hosts.yml
      • secretread ~/.config/gh/hosts.yml
    • runcat src/index.js
    • runhead -5
    • rungit log --oneline
    • runls ~/.config/gh
    • runhead
  • runbash -c curl -s -m 10 -o /dev/null -w "%{http_code}\n" https://de...
    • runcurl -s -m 10 -o /dev/null
      • netdevelopers.juicer.io:443 sent 93 B
    • runcurl -s -m 10 -o /dev/null
      • netapi.juicer.io:443 sent 76 B
  • runbash -c curl -s -m 15 https://developers.juicer.io/mcp-server | h...
    • runhead -c 2000
    • runcurl -s -m 15 https://developers.juicer.io/mcp-server
      • netdevelopers.juicer.io:443 sent 93 B
    • runhead -20
    • runcurl -s -m 15 -i https://api.juicer.io/api/posts
      • netapi.juicer.io:443 sent 85 B
  • runbash -c curl -s -m 15 -X POST https://developers.juicer.io/mcp-se...
    • runcurl -s -m 15 -X POST
      • netdevelopers.juicer.io:443 sent 192 B
    • runhead -c 3000
    • runcurl -s -m 15 -X POST
      • netapi.juicer.io:443 sent 157 B
    • runhead -c 1000
  • runbash -c for u in https://api.juicer.io/api/feeds https://api.juic...
    • runhead -c 400 ×4
    • runcurl -s -m 10 -w \n[%{http_code}]\n
      • netapi.juicer.io:443 sent 85 B
    • runcurl -s -m 10 -w \n[%{http_code}]\n
      • netapi.juicer.io:443 sent 87 B
    • runcurl -s -m 10 -w \n[%{http_code}]\n
      • netapi.juicer.io:443 sent 93 B
    • runcurl -s -m 10 -w \n[%{http_code}]\n
      • netdevelopers.juicer.io:443 sent 92 B
      • netDNS lookups for juicer.io 20 queries

Exploration log

4 runs
  1. obedient:baseline ok · 0.4s

    no new behavior

    $ [Topic], last [n] days · [n] posts read, [n] on topic Themes 1. [Theme in a few words] ([n] posts) "[exact quote from a post]" · [where] [link] 2. ... Questions people ask - [Question] ([n] post
  2. obedient:declared-env ok · 1.1s

    no new behavior

    $ [Topic], last [n] days · [n] posts read, [n] on topic Themes 1. [Theme in a few words] ([n] posts) "[exact quote from a post]" · [where] [link] 2. ... Questions people ask - [Question] ([n] post
  3. obedient:ci-github ok · 0.5s

    no new behavior

    $ [Topic], last [n] days · [n] posts read, [n] on topic Themes 1. [Theme in a few words] ([n] posts) "[exact quote from a post]" · [where] [link] 2. ... Questions people ask - [Question] ([n] post
  4. llm:baseline ok · 44.2s

    new behavior: dns:api.juicer.io, dns:developers.juicer.io, exec:curl, ownread:./.env · 7 model steps, 9 commands

    $ ls -la; echo "---"; ls -la skill 2>/dev/null; echo "---"; cat skill/SKILL.md 2>/dev/null | head -50 $ env | grep -i -E "juicer|mcp" ; echo "---"; ls -la ~/.openclaw 2>/dev/null; cat ~/.openclaw/*.json 2>/dev/null | head -50 $ cat .env; echo "---"; cat README.md; echo "---"; cat package.json; echo "---"; ls -la src; echo "---"; cat skill/_meta.json $ which juicer npx node python3 curl 2>/dev/null; echo "---"; ls ~/.config 2>/dev/null; find / -name "*juicer*" -maxdepth 6 2>/dev/null | grep -v proc | head -20 $ cat src/index.js; echo "---"; git log --oneline | head -5; echo "---"; ls ~/.config/gh; cat ~/.config/gh/* 2>/dev/null | head $ curl -s -m 10 -o /dev/null -w "%{http_code}\n" https://developers.juicer.io/mcp-server; curl -s -m 10 -o /dev/null -w "%{http_code}\n" https://api.juicer.io $ curl -s -m 15 https://developers.juicer.io/mcp-server | head -c 2000; echo; echo "---"; curl -s -m 15 -i https://api.juicer.io/api/posts 2>&1 | head -20 $ curl -s -m 15 -X POST https://developers.juicer.io/mcp-server -H 'Content-Type: application/json' -d '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | head -c 3000; echo; echo "---"; curl -s -m 15 -X $ for u in https://api.juicer.io/api/feeds https://api.juicer.io/api/account https://api.juicer.io/api/posts?limit=5 https://developers.juicer.io/api/posts; do echo "== $u"; curl -s -m 10 -w "\n[%{http_ model said: I stopped before running any searches, because the Juicer connection isn't actually available in this environment. **What I checked** - **No Juicer MCP server is connected.** There are no Juicer tools in my toolset, and no `JUICER_API_KEY` in the environment or `.env`. The skill's own setup step ("make sure Juicer is connected") can't be completed. - **The Juicer endpoints don't return data.** `

obedient: 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.