Academic Deep Research 1.0.0
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· 109·2 current·2 all-time
by@dujch
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name and description match the instructions: the skill is a procedure for rigorous, multi-cycle web research using OpenClaw native tools. No binaries, installs, or external credentials are requested, which is proportionate to the stated goal. One small inconsistency: README claims "Works offline — No API keys, no cloud services," but the runtime instructions require web_search and web_fetch (network access). This is likely a documentation mismatch rather than a functional security problem, but you should confirm the author's intent (offline vs. live web queries).
Instruction Scope
SKILL.md explicitly limits behavior to clarifying questions, presenting a research plan for approval, then executing multi-cycle web_search/web_fetch investigations and referencing memory_search/memory_get. The instructions do not ask the agent to read arbitrary system files, access unrelated environment variables, or post results to external endpoints beyond standard web fetching. Note: the skill mandates revealing detailed intermediate reasoning after every tool call, which may surface user-provided context or sensitive fragments if the user supplies them.
Install Mechanism
Instruction-only skill with no install spec and no code files to execute. Nothing is written to disk or downloaded. This is the lowest-risk install profile.
Credentials
Requires no environment variables, no credentials, and no config paths. It does use platform-provided tools (web_search, web_fetch, sessions_spawn, memory APIs) which are appropriate for web research. There are no unexplained credential requests.
Persistence & Privilege
always is false and the skill is not requesting any elevated or persistent privileges. It references memory_search/memory_get (access to agent memory) which is expected for cross-referencing prior work — confirm your agent's memory policy if you are concerned about storing outputs or sensitive inputs.
Assessment
This skill is internally coherent and uses only OpenClaw-native web and memory tools; it does not request credentials or install code. Before using it, consider: (1) Clarify the README claim about "works offline" — the skill will perform live web searches/fetches unless your environment restricts that. (2) Do not ask it to research sensitive internal data or protected health information unless you are sure your environment/network/memory policies prevent external transmission, because web_search/web_fetch will contact external sites and the skill records detailed intermediate reasoning after every tool call (which could expose sensitive fragments). (3) Expect a lot of verbosity and many external requests (count=20 per search, two cycles per theme); this can produce heavy network traffic and long outputs. If you want to use this for proprietary or regulated topics, confirm with the platform how web_fetch results are handled and whether you can disable memory writes or restrict network access.Like a lobster shell, security has layers — review code before you run it.
latestvk971xc6e0dhm44a9gxnk006fys839wc9
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🔬 Clawdis