Caveat verdict

ultra-memory

88
🟢 Trusted
No high-risk patterns surfaced by the deep scan — automated capability review, not behavioral proof.

Cross-session agent memory skill that stores data locally via bundled Python scripts; network_in is for optional package installation and the skill explicitly describes local-only storage with no outbound exfiltration.

⚠ Flagged for review — coarse, uncorroborated signal, not a confirmed exploit. Review the config yourself before installing.

Automated static analysis — not a human review. Caveat flags capabilities, not confirmed intent, and can produce false positives. Disagree with this verdict? Use Dispute below.

0
security
60
transparency
80
maintenance

Permission integrity

Installs packages at runtime — transitive dependencies are not auditable

package_install

Accesses agent memory/configuration files

agent_memory

Findings (18)

Pattern match critical

Accesses sensitive system files

references/advanced-config.md · code · /etc/passwd

Pattern match high

Pipe to python — executes piped content as Python code

CLAWHUB.md · prose · downgraded · | Python

Pattern match high

subprocess execution — runs system commands from Python

references/advanced-config.md · code · subprocess.run(

Pattern match high

HTTP request to bare IP address — common in malicious payloads

platform/SYSTEM_PROMPT.md · code · http://127.0.0.1

Pattern match high

Accesses .ssh directory

references/advanced-config.md · code · .ssh/

Pattern match high

References sudo — requests elevated privileges

references/advanced-config.md · code · sudo

Pattern match high

Pipe to sh — executes piped content as shell commands

scripts/detect_contradictions.py · prose · downgraded · |sh

Pattern match high

Pipe to bash — executes piped content as shell commands

scripts/hook_capture.py · prose · downgraded · |Bash

Pattern match medium

References agent memory files

SKILL.md · code · memory.md

Pattern match medium

References child_process — can spawn system processes

scripts/mcp-server.js · prose · downgraded · child_process

Pattern match low

Python os.environ.get — reads environment variable

integrations/langchain_memory.py · prose · downgraded · os.environ.get(

Pattern match low

References agent configuration files

references/advanced-config.md · prose · downgraded · CLAUDE.md

Pattern match low

Sets world-executable permissions

references/advanced-config.md · prose · downgraded · chmod 700

Pattern match low

Python shutil file operation — copies/moves/deletes files

scripts/cleanup.py · prose · downgraded · shutil.move(

Pattern match low

Popular HTTP library — network access

scripts/log_op.py · prose · downgraded · axios

Pattern match low

pip3 install — installs Python packages at runtime

scripts/log_op.py · prose · downgraded · pip3 install

Pattern match low

conda install — installs packages via conda

scripts/log_op.py · prose · downgraded · conda install

Pattern match low

Python urllib.request — network access

test_e2e.py · prose · downgraded · urllib.request

Why the tier is capped

Execution sink present in raw bytes (Hard Floor: class A/D). Final tier capped at Caution — cannot be lifted by any downgrade, example-payload opt-in, or allowlist.

Permissions & capabilities

No declared permissions — minimal attack surface.

package_installagent_memorynetwork_in

Is this flag fair?

Check another skill Browse the registry Auditing your own skills or configs? Use the API