• Current research work (hopefully PhD): Muti-agent Simulation, MedVoice.
• Open source contributor: OpenBMB/ChatDev 2.0, HKUDS/RAG-Anything.

Auto-synced daily from my-hermes-skills · docs

Research

• covidence-full-text-retrieval — Autonomously retrieve full-text PDFs for references in the full-text review stage of a Covidence systematic review, in the user's Chrome session via CDP. Looks up open-access copies via Unpaywall, Semantic Scholar, and arXiv, with an optional NotebookLM-assisted last-resort web search, then uploads any PDF it finds through Covidence's "Upload full text" action. References with no locatable full text get a note logged for manual follow-up. Never casts an Include/Exclude decision -- that stays manual. (v1.2.0)
• covidence-full-text-review — Review full-text papers in a Covidence systematic review. Mode 1 (websearch_queue): walks the Covidence queue, web-searches each title, opens the first result in the papers-access browser profile, reads the full text, and reports verdicts in chat (no Covidence writes). Mode 2 (secondary_reviewer): receives a URL from the user, reads the paper, returns a structured Include/Exclude verdict with rationale, and optionally casts the vote in Covidence when a ref_id is supplied. (v2.0.0)
• covidence-screening — Autonomously screen references at the title & abstract stage of a Covidence systematic review in the user's Chrome session via CDP. Votes Yes/Maybe/No per the user's PICO criteria, writes a one-line rationale for Maybe votes, and runs unattended after an approve-first-N onboarding phase. (v1.0.0)
• sci-hub-access — Use when a paywalled DOI's full text is needed and no open-access copy exists — retrieving PDFs via Sci-Hub mirrors, and joining/using Sci-Net (sci-net.xyz) for post-2022 papers. Covers mirror discovery, direct PDF fetch, Turnstile gate, and the Sci-Net invite-code API. (v1.1.0)

Productivity

• auditing-worktrees — Use when asked for the status of worktrees, which worktree branches are ready to review, pushed, merged or shipped, which worktrees are stale, idle or purgeable, or for a periodic worktree audit across one or several repos (v1.0.0)
• hermes-deepseek-cache — Use when running Hermes Agent with a DeepSeek-family model (deepseek, or deepseek-v4-flash/pro via an aggregator like opencode-go) and want to see, graph, or reset DeepSeek's prefix-cache hit rate, diagnose cache misses, or confirm the deepseek_cache.enabled config — OR when you want to port/apply the deepseek_cache.py module (DeepSeek prefix-cache telemetry, /cache-stats /cache-graph /cache-reset, prefix-break diagnostics, stable tool ordering, cache-friendly compaction) into a Hermes source tree. Covers use + integration. (v1.1.0)
• i-have-adhd — Shape output for a reader with ADHD: lead with the next action, number multi-step work, restate state across turns, suppress tangents, give specific time estimates, make wins visible. Invoke with /i-have-adhd; stays on until "stop adhd mode".
• memory-ui — Launch the Mnemosyne browser UI to inspect stored memory.
• oh-my-ghostty — Use when setting up or refreshing the oh-my-ghostty terminal stack on macOS - Ghostty config, fastfetch on new-terminal startup, and bat/eza/zoxide aliases for zsh. (v1.0.0)
• tldr — Use when the user says tldr / TL;DR or asks for a before/after summary. (v1.1.0)
• writing-reports — Use when the user asks for a report, research write-up, benchmark/model/tool comparison, or any 'give me the report' request. (v1.0.0)

Coding Agents

• oh-my-pi — Troubleshoot/operate the Oh-My-Pi (omp) coding agent.

Slack

• slack-mention-watch — Use when the user asks about Slack pings, mentions, notifications or what they missed, or wants @mention alerts or a triaged mention digest while away. Read-only. (v2.2.0)
• slack-scan — Summarize today's Slack channel activity on demand (bot-member channels). NOT for pings/mentions -- use slack-mention-watch. (v1.0.0)

ilearning-autoadvance

• ilearning-autoadvance — Autonomously advance HCL iLearning / SAP SuccessFactors compliance courses (Articulate Storyline SCORM, Ethena GoEthena, or direct-HTML SVG players) via raw CDP on port 9222. Resumes paused audio, clicks interactive slide topics/cards, answers quizzes, navigates to the next slide. (v5.0.0)

superpowers

• superpowers-using-superpowers — Use when starting any conversation - establishes how to find and use skills, requiring skill loading (skill_view) before ANY response including clarifying questions (v1.0.0)