Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Process Flow Navigator
v1.0.0业务流程图导航助手 - 帮助你在复杂的多分支流程中导航、规划路径、查询技能编码。支持 A→K 主流程及所有子流程分支。
⭐ 0· 210·0 current·0 all-time
by@jopyin
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The skill's name/description (process flow navigation) match the included files: a JSON flow-rules data file, SKILL.md, README, and a CLI script that answers navigation queries. One minor inconsistency: the SKILL.md/registry metadata declare no required binaries, but scripts/navigate.sh invokes python3 for the 'code' command. That is a small metadata mismatch (the script legitimately needs python3 to print node codes) but does not indicate malicious behavior.
Instruction Scope
SKILL.md describes navigation, path planning, and code lookup only. The runtime script reads only the included data/flow-rules.json and prints results; it does not reference arbitrary user files, system configuration, or network endpoints. The SKILL.md notes the source was a user-provided PDF, but there is no code that tries to parse user PDFs or automatically ingest user files, so there is no scope creep.
Install Mechanism
No install spec is present (instruction-only plus a local script); nothing is downloaded at install time. README suggests git clone or clawhub install but there are no remote installers or external archives referenced, so no high-risk install behavior is present.
Credentials
The skill declares no environment variables, no credentials, and none of the files attempt to read environment secrets. All data used is in the included data/flow-rules.json; there is no disproportionate credential request.
Persistence & Privilege
always is false and the skill does not request persistent system-wide changes or modify other skills. It is a normal, non-persistent helper with no special privileges.
Assessment
This skill appears coherent and local-only, but exercise the usual caution: (1) the CLI script uses python3 though the metadata lists no required binaries — ensure python3 is available or adapt the script before use; (2) the source/homepage are unknown, so review data/flow-rules.json to confirm the flow matches your expectations and contains no sensitive information you wouldn't want stored in a skill; (3) run the script in a sandbox or non-production environment first if you have any doubts; (4) because the skill is user-invocable and can be invoked by the agent autonomously (normal default), only enable it if you trust the owner or have inspected the files — there is no evidence of network calls or credential access in the package.Like a lobster shell, security has layers — review code before you run it.
latestvk97646vj484vj4syatk69v83dd82pgj3
License
MIT-0
Free to use, modify, and redistribute. No attribution required.