Systematic Review Screener
v1.0.0Automated abstract screening tool for systematic literature reviews with PRISMA workflow support.
⭐ 0· 39·0 current·0 all-time
byAIpoch@aipoch-ai
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description, SKILL.md, sample references, criteria template, and scripts/main.py align: all artifacts are consistent with an abstract screening / PRISMA export tool. No unrelated binaries, env vars, or services are requested.
Instruction Scope
SKILL.md instructs running the packaged script with local inputs and to verify inputs/paths first. It does not ask the agent to read unrelated system files or secrets. Note: SKILL.md and the script emphasize manual verification and reproducible outputs — appropriate for this domain.
Install Mechanism
Instruction-only with no install spec; code is included in-repo (no downloads). Dependencies are limited (PyYAML) and requirements.txt only lists dataclasses and yaml. No installer or external download URLs observed.
Credentials
The skill requires no environment variables, credentials, or config paths. The declared requirements match the task (local file inputs and YAML criteria). No secret-exfiltration indicators in the provided files.
Persistence & Privilege
always:false and no code attempts to modify agent/system settings or other skills. The script reads/writes local input/output files as expected for a screening tool.
Assessment
This skill appears coherent and self-contained for screening abstracts, but treat it like any third-party code: 1) Inspect scripts/main.py fully before running (it’s included) and run python -m py_compile scripts/main.py as suggested to detect syntax errors. 2) Execute first on a small sample dataset in an isolated environment (VM or container) to verify behavior and outputs. 3) The requirements include PyYAML (install from a trusted source); dataclasses is built into modern Python and is likely unnecessary. 4) Do not feed sensitive/private data (PHI) until you have validated that outputs are correct and no unexpected logging/telemetry occurs. 5) Expect there may be minor bugs (I noticed an apparent variable typo in the CSV parser in the provided truncated code); be prepared to patch or contact the author. If you want, I can (a) run a more detailed static review of the full main.py for logic errors, or (b) suggest minimal fixes for the CSV parsing/runtime issues.Like a lobster shell, security has layers — review code before you run it.
latestvk9776f5d6jcr75zw9876sg5yyn841vgg
License
MIT-0
Free to use, modify, and redistribute. No attribution required.