Deepxiv Cli

v0.2.2

Search, inspect, and progressively read open-access academic papers with the deepxiv CLI. Use when the user wants arXiv / PMC / Semantic Scholar paper search...

0· 86·0 current·0 all-time
byEric@xupeng8
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Pending
View report →
OpenClawOpenClaw
Benign
high confidence
Purpose & Capability
Name/description (CLI for searching and progressively reading open-access papers) matches the instructions and referenced commands (search, paper, pmc, sc, trending, wsearch, agent). No unrelated credentials, binaries, or config paths are requested.
Instruction Scope
SKILL.md stays on-topic: it tells the agent to check for deepxiv, run its help/health/debug, and to follow progressive-reading workflows. It explicitly forbids installing without user approval and documents LLM-agent caveats. There are no steps that read unrelated files or exfiltrate data.
Install Mechanism
There is no automatic install spec in the skill bundle (instruction-only). references/install.md directs users to install via pipx or pip --user and to use system package managers (apt/dnf/brew/pacman) for pipx. This is expected for a Python CLI but does involve installing packages and (in some cases) sudo; user consent is required before running those commands.
Credentials
The skill declares no required environment variables or credentials. It does require the user's LLM backend to be configured if using the `agent query` feature, and SKILL.md notes that LLM usage will consume the user's account — this is documented and proportional to the feature.
Persistence & Privilege
always is false and the skill does not request persistent/system-wide modifications. It does not attempt to change other skills' configs or require elevated persistent privileges.
Assessment
This skill is an instruction-only helper for an external Python CLI (deepxiv). Before installing: (1) confirm you want the deepxiv-sdk Python package from PyPI and be prepared to run pipx or pip install commands (some platforms use sudo); the skill explicitly says the agent must ask you before installing anything — do not let the agent install without approval. (2) If you plan to use the `agent query` LLM feature, understand it will use your LLM account and incur usage/costs; configure the LLM backend yourself. (3) Consider verifying the deepxiv-sdk package source (PyPI page, project homepage or repo) and check reviews or repository before installing. (4) If you prefer not to install system packages or grant sudo, decline installation and ask the agent to proceed only with commands that do not require installing deepxiv. Overall the skill is coherent, but exercise the usual caution when running package installs and when enabling LLM-backed autonomous features.

Like a lobster shell, security has layers — review code before you run it.

latestvk97azdd497dy926wnv20yp7pqd84gzh5

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments