Deep Current
v2.0.0Persistent research thread manager with a CLI for tracking topics, notes, sources, and findings. Pair with a nightly cron job to build a personal research di...
⭐ 0· 225·0 current·0 all-time
byMei Park@meimakes
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (research thread manager) match the included files: a zero-dependency Python CLI and instruction prompts that use the agent's web_search/web_fetch tools. The required binary (python3) and file read/write access to deep-current and deep-current-reports are appropriate for the stated functionality.
Instruction Scope
SKILL.md instructs the agent to pick threads, use its web_search/web_fetch tools to research, and write reports to deep-current-reports/YYYY-MM-DD.md — that scope is consistent with the skill's purpose. The runtime instructions do not request unrelated files, credentials, or external endpoints beyond the agent's own web tools.
Install Mechanism
No install spec is provided (instruction-only), and the shipped code is included in the skill bundle. Nothing is downloaded from external URLs and no archives are extracted, which is the lower-risk pattern.
Credentials
The skill requires only python3 and no environment secrets. Metadata grants file read/write within a workspace area which matches purpose, but there is a small naming mismatch: SKILL.md/metadata mention deep-current-threads while the code expects/creates 'deep-current' and 'deep-current-reports', and the script prefers a workspace path under ~/.openclaw/workspace/deep-current. This is likely benign but you should confirm the target directories before running.
Persistence & Privilege
The skill does not request always:true, does not modify other skills, and only persists its own data file (currents.json) and report files. Agent autonomous invocation is the platform default and is not in itself a concern here.
Assessment
This skill appears to do what it says: a local Python CLI storing data in a workspace and prompting the agent to use its web_search/web_fetch tools to create nightly reports. Before installing or scheduling autonomous runs: 1) Inspect the included script (scripts/deep-current.py) — it reads/writes ~/.openclaw/workspace/deep-current/currents.json (or a skill-relative deep-current/currents.json) and writes reports to deep-current-reports/; back up any existing data at those paths. 2) Confirm where you want data stored and adjust the cron prompt or file locations if needed (metadata references deep-current-threads which is inconsistent with the script). 3) Understand the agent will perform web searches and write files — ensure your agent's web tools and cron environment have the appropriate network and file permissions and avoid giving the agent secrets in the cron prompt. 4) Because the skill will run locally and write to your home/workspace, review and test the CLI manually (python3 scripts/deep-current.py list/add/note) before enabling automated nightly runs. No regex scan findings were flagged in the package.Like a lobster shell, security has layers — review code before you run it.
latestvk97c3c7wvp77mzbw4jaezzc4bx83ggj3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
Binspython3