paperclip

SAFE max severity MEDIUMSkillSpector Do not install 100 registry K-Dense-AI last scanned 1 scan

Verdict safe, but 2 findings (1 medium, 1 low), mainly data exfiltration. Most recent scan 2026-10-21.

Findings2
Critical0
High0
Medium1
Low1
Analyzersstatic bytecode pipeline correlation behavioral trigger
Scanner version2.3.1

Second scanner: NVIDIA SkillSpector

VerdictDo not install 100
Score100 / 100
Issues13
CategoriesPrivilege Escalation Tool Misuse Data Exfiltration
Version2.12.0 (static)

The two scanners disagree: Cisco found nothing and SkillSpector says do not install. One of them has a blind spot here; read the skill before trusting either. How the two scanners compare across the whole corpus: Two scanners, one corpus.

What this skill does

Searches and reads biomedical papers, FDA/PMDA/EMA documents, clinical trials, and protein records with the GXL Paperclip CLI and Python SDK. Supports source-scoped search, full-text grep, metadata SQL, map/reduce extraction, figure analysis, optional repositories and claim verification, and line-pinned citations. Use when a task names GXL paperclip, asks to install or authenticate it, or requests literature retrieval and evidence extraction through Paperclip.

Package
7 files, 70 KB
SKILL.md
1546 words, 12+ sections
Code
None. Instructions only.
Allowed tools
Bash Read Write
License
MIT
Compatibility
Requires network access and the GXL paperclip CLI. The macOS/Linux installer requires Python 3.8+ plus curl or wget; it installs a Python launcher and private library, not an interpreter. Use PAPERCLIP_API_KEY or existing browser-login credentials. Hosted MCP is available without local CLI installation. Reviewed against CLI/SDK 0.7.92.

SKILL.md covers: Paperclip CLI · Preflight and authentication · Choose the retrieval operation · Find, read, and cite · Multi-paper extraction · Substitute the actual search ID below. · Regulatory, trial, and figure reads · Use the actual filename returned by ls. · Virtual filesystem · Citations · Routines and repositories · Version boundaries

Findings

MEDIUM Python code block sends HTTP POST request Data Exfiltrationbehavioral

Code block in references/python-sdk.md at line 283 contains potentially dangerous Python code.

references/python-sdk.md:283

response = requests.post(
Remediation: Review the code block for security implications.
LOW Dangerous data flow in command pipeline Data Exfiltrationpipeline

Pipeline downloads data from the network and executes it: `curl -fsSL https://paperclip.gxl.ai/install.sh | bash`. This is a remote code execution pattern. (Note: found in documentation file - may be instructional rather than executable.)

references/installation.md:1

curl -fsSL https://paperclip.gxl.ai/install.sh | bash
Remediation: Review the command pipeline. Avoid piping sensitive data to network commands or shell execution.

What is in the package

FileKindSize
SKILL.mddoc13 KB
references/cli-reference.mddoc9.3 KB
references/installation.mddoc7.3 KB
references/map-reduce.mddoc7.0 KB
references/python-sdk.mddoc16 KB
references/repos-and-workspace.mddoc8.6 KB
references/search-and-retrieval.mddoc9.1 KB

External hosts referenced

paperclip.gxl.ai arxiv.org doi.org export.arxiv.org

Hostnames that appear in the skill's files. Referencing a host is not a finding; compare against the findings above and the skill's stated purpose.

Environment variables referenced

PAPERCLIP_API_KEY PAPERCLIP_BEARER_TOKEN PAPERCLIP_BASE_URL

Package contents recorded 2026-10-22 at commit 593377bb0487. The scan above ran against the version current at scan time; the two can differ if the skill has since been updated.

What we checked

What we did not check

Sources

Related skills