ClawHub

七宗罪.skill / 7 Sins.skill

v1.0.3

用七宗罪审判产品、功能、商业构想与人生选择，撕掉包装，只留下欲望、代价与真相。

⭐ 1· 47·0 current·0 all-time
bySHAO Zhaoru@shaozrrr
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
Capability signals
CryptoCan make purchases
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
high confidence
✓
Purpose & Capability
The skill's name and description (a seven-vice, adversarial product/decision review) match the SKILL.md instructions: a parser + seven biased agents + aggregator. It does not request unrelated binaries, credentials, or config paths. The agent definitions mention web_search for some sin-agents, which is consistent with the goal of pulling public evidence for monetization/competitive claims.
✓
Instruction Scope
All runtime instructions are about parsing user input, detecting mode, running three rounds of biased evaluations, and producing a fixed five-section visible output. The SKILL.md does not instruct reading local files, environment variables, or sending data to unexpected endpoints. It does instruct that some agents use web search; that implies external network queries only if the host runtime exposes a search connector.
✓
Install Mechanism
No install spec and no code files beyond prompt/orchestration materials. This is the lowest-risk form: nothing is written to disk or fetched by the skill itself.
✓
Credentials
The skill declares no required environment variables, credentials, or config paths. The only external capability referenced is web_search for evidence gathering — reasonable for its purpose but dependent on host-provided connectors (not declared as env creds in the skill).
ℹ
Persistence & Privilege
always:false (normal). agents/openai.yaml sets policy.allow_implicit_invocation: true, which may allow the host to invoke this skill implicitly in some environments; by itself this is not high privilege and is proportional given the skill's purpose, but you should be aware the host platform may surface or auto-run skills depending on its own policies.
Assessment
This skill is an instruction-only prompt bundle (no binaries, no install, no secrets) and appears internally consistent with its goal of adversarial product/decision review. Before installing: 1) Check whether your host/runtime exposes a web-search connector — if it does, the skill's Greed/Envy agents will attempt web lookups and those queries (including your prompt text) could be sent to the search provider; avoid sending proprietary or sensitive data. 2) Note allow_implicit_invocation is enabled in the interface metadata — depending on your platform that may allow the skill to be suggested or auto-invoked; if you prefer only explicit use, verify platform settings. 3) The skill enforces a strict visible output format (five headings, table rules); if you need a different output shape, modify prompts. 4) If you expect automated scripts or integrations, this skill doesn't ship any — it's purely prompt orchestration.

Like a lobster shell, security has layers — review code before you run it.

latestvk97f5tftrsdn9fqtnzqn49kkjs84s9dy

License

MIT-0
Free to use, modify, and redistribute. No attribution required.
Termshttps://spdx.org/licenses/MIT-0.html

Comments