deep-currentPersistent 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...
Install via ClawdBot CLI:
clawdbot install meimakes/deep-currentGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
report → https://github.com/meimakes/deep-currentCalls external URL not in known-safe list
https://github.com/meimakes/deep-currentAI Analysis
The skill is a local Python CLI for managing research threads as JSON files. The external URL (GitHub) is the project's homepage, not a data sink. All data operations are confined to the user's workspace directories, and research is performed using the agent's standard, user-controlled tools.
Audited Apr 17, 2026 · audit v1.0
Generated Oct 6, 2026
A market analyst subscribes Deep Current to a nightly cron job that tracks competitor pricing pages, product launches, and funding announcements across five rival companies. Each morning they receive a dense digest in deep-current-reports/ with fresh sources and actionable flags. Over weeks the threads accumulate context that a single search session could never match.
An early-stage investor maintains persistent threads on emerging niches like AI agents and climate fintech, letting the nightly agent surface new startups, papers, and fundraises. The covered command prevents duplicate sources across weeks, keeping the digest high-signal. Thread lifecycle status helps park sectors that have cooled.
A researcher tracks several subfields with separate threads, using the nightly job to pull new preprints, citations, and blog discussions. Notes and findings accumulate into a personal annotated bibliography, while the decay command prunes topics no longer active. Reports become a searchable archive in markdown.
A reporter follows multiple beats via threads (local policy, tech regulation, housing) and lets the agent cross-reference sources and flag actionable leads each night. The dense report format is written for a smart reader, making it easy to scan before morning editorial meetings. The covered command ensures the same story angle isn't repeated.
A curious professional runs Deep Current on non-work interests like specialty coffee, urbanism, or retro computing, building a slow-burn research journal. Each nightly digest offers something engaging to read over coffee, with thread titles written to be catchy. Findings accumulate with inline links, forming a personal archive.
The Python CLI remains free and self-hosted, but a paid tier offers a managed cron service that runs the nightly research and emails or hosts the reports. Users avoid setting up their own scheduler and model budget while retaining thread portability.
A collaborative layer that shares threads, notes, and digests across an analyst team, with role-based access, shared source pools, and weekly rollup reports. Targets consultancies, newsrooms, and VC research teams that need institutional memory rather than a single user's journal.
Run public-facing Deep Current threads on trending sectors and publish the resulting digests as a paid newsletter or research briefing. The skill becomes the production tool for an audience-facing product rather than an internal utility.
💬 Integration Tip
Set up an isolated nightly cron job with a capable model and a 30-minute timeout, and run the covered command first each session to avoid repeating recent sources. Keep threads narrowly scoped so each digest stays dense and the decay command can prune genuinely stale topics.
Scored Oct 6, 2026
Create, list, modify, and remove scheduled cron jobs to automate system tasks using simplified cron syntax and manage output logging.
Manage n8n workflows and automations via API. Use when working with n8n workflows, executions, or automation tasks - listing workflows, activating/deactivating, checking execution status, manually triggering workflows, or debugging automation issues.
Diagnose and triage cron job failures. Checks job states, identifies error patterns, prioritizes by criticality, generates health reports. Triggers on: cron...
Decomposes complex user requests into executable subtasks, identifies required capabilities, searches for existing skills at skills.sh, and creates new skills when no solution exists. This skill should be used when the user submits a complex multi-step request, wants to automate workflows, or needs help breaking down large tasks into manageable pieces.
Set reminders using natural language. Automatically creates one-time cron jobs and logs to markdown.
Performs a two-phase audit combining a fast deterministic scan and a deep LLM quality review of security, cron jobs, config, and skills.