Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
ideas
v1.0.0记录日常奇思妙想、灵感片段、技术创意。支持按主题分类、状态追踪、关键词搜索。
⭐ 0· 54·0 current·0 all-time
byKazuya@kazuya-ecnu
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
high confidencePurpose & Capability
The skill claims to parse content, extract tags, generate IDs, append entries, search, update statuses, and so on. The included agent.py contains only trivial read/write helper functions and a print message — it does not implement tag extraction, ID generation, searching, filtering, or the data path declared in SKILL.md. This mismatch suggests the code and the stated capability are inconsistent.
Instruction Scope
SKILL.md instructs the agent to read/write data at ~/.openclaw/workspace/ideas-data/ideas.json and describes detailed operations (parse content, extract tags, generate IDs, etc.). The code, however, references a different path (~/.openclaw/skills/ideas/ideas.json) and does not perform the described operations. While all file IO is local (no network/external endpoints), the conflicting paths and the lack of implemented logic are scope/integration inconsistencies.
Install Mechanism
No install spec is provided (instruction-only plus a small helper script). This is low-risk from an install/download perspective — nothing is fetched from external URLs and no installers run.
Credentials
The skill requests no environment variables, credentials, or external config paths. All described data storage is under the user's home directory. There are no signs of unrelated credential access or external endpoints.
Persistence & Privilege
always is false and the skill does not declare elevated privileges or manipulate other skills' configurations. It would store data locally under a path in the user's home if used.
What to consider before installing
This skill appears to be a simple local idea tracker, but there are important inconsistencies you should resolve before trusting it with your notes:
- Path mismatch: SKILL.md says data lives in ~/.openclaw/workspace/ideas-data/ideas.json, while agent.py writes to ~/.openclaw/skills/ideas/ideas.json. Decide which path is authoritative and fix the mismatch so your data isn't lost or split.
- Missing implementation: SKILL.md describes features (tag extraction, ID generation, search, status updates) but agent.py doesn't implement them — the AI runtime would need to perform those steps. If you expect those features to work offline/locally, verify where the logic runs and whether the agent has permission to read/write the intended path.
- Data locality: All storage is under your home directory. If you have privacy concerns, inspect the code and SKILL.md, and confirm where files will be written. Consider backing up the directory or changing the path to a location you control.
If you want to proceed: either update agent.py to implement the SKILL.md behavior and use a single, explicit data path, or adjust SKILL.md to reflect the actual runtime behavior. If you are unsure, do not install it system-wide until these inconsistencies are resolved.Like a lobster shell, security has layers — review code before you run it.
latestvk977b6bpv85np3cnt1f54rpved8484mc
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
