Related Works Report from Paper MDs

Name/description match the included assets: SKILL.md describes extraction, deduplication, and abstract lookup; the repo includes parsing and assembly scripts that implement those tasks. There are no unrelated binaries, env vars, or config paths requested.

Instructions stay within the stated workflow: read provided paper markdowns, write all artifacts under a user-selected WORKDIR, run title-to-arXiv matching via Tavily, and use local helper scripts to fetch and render abstracts. Two notes: (1) the SKILL mandates 'Tavily-only' for title lookups and 'no fallback' — this is a functional constraint (not a covert action) but may result in partial results if Tavily fails; (2) using Tavily (an external MCP/search) means paper titles (and any text included in queries) will be sent to that third party for matching — expected for this task but worth user awareness.

No install spec — instruction-only skill with bundled helper scripts. No downloads or archive extraction. The included Python scripts are small, readable, and operate locally or contact arXiv; nothing is written to disk outside the user-specified WORKDIR per the instructions.

The skill requests no environment variables or credentials (proportionate). Caveat: the workflow depends on a third-party service (Tavily) for title->arXiv matching; while the skill doesn't request secrets, using Tavily will transmit titles/queries to that service. The included fetch script contacts only arxiv.org for canonical abstracts/metadata.

always is false, autonomous invocation is allowed (platform default). The skill does not request permanent system-wide privileges or modify other skills/config. It writes only into the chosen WORKDIR.