Academic Deep Research
v1.0.0Transparent, rigorous research with full methodology — not a black-box API wrapper. Conducts exhaustive investigation through mandated 2-cycle research per t...
⭐ 0· 59·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (deep, reproducible research) align with the runtime instructions: heavy use of platform web_search, web_fetch, sessions_spawn, and explicit multi-cycle methodology. No unrelated binaries, credentials, or installs are requested.
Instruction Scope
Instructions stay focused on research tasks and demand explicit check-points and reproducibility. Two minor notes: (1) the README claims 'Works offline' while the SKILL.md mandates web_search/web_fetch (internet use) — this is a documentation inconsistency; (2) the skill instructs use of memory_search/memory_get to cross-reference prior context (expected for deep research) — users should be aware this will read agent memory which may contain sensitive material.
Install Mechanism
Instruction-only skill with no install spec, no downloads, and no declared external package requirements — lowest-risk install footprint.
Credentials
No environment variables, credentials, or config paths are requested. The only potentially sensitive access is to agent memory (memory_search/memory_get) and to external web sources via web_fetch/web_search — both are coherent with research functionality but warrant user awareness.
Persistence & Privilege
Skill does not request always:true and uses default agent invocation behavior. It requires user approvals at Phase 1 and Phase 2 before full execution (Phase 3 runs automatically after that approval) — this is proportionate to its purpose.
Assessment
This skill is internally coherent and appears to do what it says: systematic web-based research using the platform's web_search/web_fetch and memory tools. Before installing, confirm two practical items: (1) privacy expectations — the skill will read agent memory if available and perform many web fetches (count=20 per search) which may transmit query text and intermediate findings to external sites; avoid providing PHI or other sensitive data unless you trust platform policies; (2) documentation mismatch — README's 'Works offline' claim conflicts with required web_search/web_fetch usage, so verify whether offline operation is actually supported in your environment. Consider testing the skill on a non-sensitive topic first and review SkillBoss tool permissions and memory contents before use.Like a lobster shell, security has layers — review code before you run it.
academicvk97b1fjkqe0cpjyhndc9ww2mpx84q0hhcitationsvk97b1fjkqe0cpjyhndc9ww2mpx84q0hhlatestvk97b1fjkqe0cpjyhndc9ww2mpx84q0hhliteraturevk97b1fjkqe0cpjyhndc9ww2mpx84q0hhresearchvk97b1fjkqe0cpjyhndc9ww2mpx84q0hh
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🔬 Clawdis