BaZi Partner 八字搭档匹配
v1.0.1Chinese BaZi (Four Pillars of Destiny) analysis and AI partner matching. Trigger when user asks about 八字, 命理, 四柱, birth chart analysis, BaZi partner, or want...
⭐ 0· 38·0 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 (BaZi analysis and AI partner matching) match the included code and data: there are algorithms to compute Four Pillars, Ten Gods, pattern L1→L2 logic, partner mapping/prompts, and an MCP server exposing bazi_analyze, bazi_partner, and bazi_apply_prompt. Required binary is only python3, which is appropriate.
Instruction Scope
SKILL.md instructs an automatic three-step flow (analyze → derive L2 → partner match → ask before writing). The code implements these tools. The important scope-related behaviour: the skill reads/writes the user's SOUL.md (~/.openclaw/SOUL.md or workspace/SOUL.md) to append the partner's system_prompt. That file access is documented in SKILL.md and consistent with the stated purpose, but it is a material effect on the user's local configuration so the agent must show the prompt and obtain explicit confirmation before calling bazi_apply_prompt as instructed.
Install Mechanism
Registry has no formal install spec; SKILL.md / README recommend pip install from the author's GitHub (standard release host). pyproject lists reasonable dependencies (mcp[cli], lunar-python). No downloads from suspicious hosts, no obfuscated installers; installing arbitrary code from GitHub is normal but should be done only if you trust the repo.
Credentials
The skill requests no environment variables or external credentials. It performs only local computation and local file operations (reads/writes SOUL.md). That local file access is proportionate to the feature (applying a persona), but it is the only sensitive I/O to be aware of.
Persistence & Privilege
always:false and normal autonomous invocation settings. The notable privilege is file modification: bazi_apply_prompt will create/modify ~/.openclaw/SOUL.md (it uses explicit markers to replace prior bazi-partner sections). This is expected behaviour for injecting a persona, but it's persistent (survives restarts) and therefore worth confirming with the user before accepting writes. Autonomous agent invocation combined with a write-capable tool increases impact if misused; however nothing in the package forces silent writes—both SKILL.md and the code instruct to ask for explicit user confirmation.
Assessment
This skill appears to do what it says: compute BaZi charts and match an AI "partner" persona, and it will offer to append the partner's system_prompt into your SOUL.md (~/.openclaw/SOUL.md). Before installing or using it: 1) Review the partner's system_prompt text returned by bazi_partner before you confirm writing — that prompt will influence the assistant's behavior. 2) Back up ~/.openclaw/SOUL.md (if present) so you can restore prior content. 3) Confirm that the agent UI actually asks you for permission before calling bazi_apply_prompt; if you prefer, decline the write and only view the suggested system_prompt. 4) If you install via pip from the GitHub repo, ensure you trust the repository/author (pip-installing code executes third-party code on your machine). 5) No network credentials or external endpoints are required by the skill, which reduces remote-exfiltration risk, but always inspect the system_prompt contents for any unexpected instructions or external URLs before approving persistence.Like a lobster shell, security has layers — review code before you run it.
latestvk978kmm72aejmnj0c6npr83zeh840n07
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🎴 Clawdis
Binspython3