← 返回 Skills 市场

AI/ML API LLM + Reasoning for OpenClaw

Name: AI/ML API LLM + Reasoning for OpenClaw
Author: aimlapihello

作者 AI/ML API · GitHub ↗ · v1.0.1

cross-platform ⚠ suspicious

1314

总下载

当前安装

版本数

在 OpenClaw 中安装

/install aiml-llm-reasoning

功能描述

使用说明 (SKILL.md)

AIMLAPI LLM + Reasoning

Overview

Use run_chat.py to call AIMLAPI chat completions with retries, optional API key file fallback, and a User-Agent header on every request.

Quick start

export AIMLAPI_API_KEY="sk-aimlapi-..."
python3 {baseDir}/scripts/run_chat.py --model aimlapi/openai/gpt-5-nano-2025-08-07 --user "Summarize this in 3 bullets."

Tasks

Run a basic chat completion

python3 {baseDir}/scripts/run_chat.py \
  --model aimlapi/openai/gpt-5-nano-2025-08-07 \
  --system "You are a concise assistant." \
  --user "Draft a project kickoff checklist." \
  --user-agent "openclaw-custom/1.0"

Add reasoning parameters

python3 {baseDir}/scripts/run_chat.py \
  --model aimlapi/openai/gpt-5-nano-2025-08-07 \
  --user "Plan a 5-step rollout for a new chatbot feature." \
  --extra-json '{"reasoning": {"effort": "medium"}, "temperature": 0.3}'

Structured JSON output

python3 {baseDir}/scripts/run_chat.py \
  --model aimlapi/openai/gpt-5-nano-2025-08-07 \
  --user "Return a JSON array of 3 project risks with mitigation." \
  --extra-json '{"response_format": {"type": "json_object"}}' \
  --output ./out/risks.json

References

references/aimlapi-llm.md: payload and troubleshooting notes.
README.md: changelog-style summary of new instructions.

安全使用建议

This skill appears to do exactly what it claims: a small helper to call AIMLAPI chat completions. Before installing/use: 1) verify the code if you can (source and homepage are missing and embedded metadata mismatches the registry), 2) only provide a dedicated AIMLAPI API key (use a least-privilege key and rotate if necessary), 3) avoid using --apikey-file with paths that contain other secrets, and 4) don't include private data in --extra-json since that is sent to the remote API. If you need stronger assurance, ask the publisher for a canonical homepage or repository and confirm the ownerId/version mismatch is intentional.

功能分析

Type: OpenClaw Skill Name: aiml-llm-reasoning Version: 1.0.1 The skill is classified as suspicious due to potential local file read/write vulnerabilities in `scripts/run_chat.py`. The `--apikey-file` and `--output` arguments allow the script to read from and write to arbitrary file paths on the system, which could be exploited via prompt injection against the OpenClaw agent to achieve local file inclusion (LFI) or local file write (LFW). While the `README.md` file explicitly warns about the `--apikey-file` flag, the capability itself presents a significant risk. There is no evidence of intentional malicious behavior (e.g., data exfiltration, persistence, or obfuscation) within the script, making it a vulnerability rather than outright malware.

能力评估

ℹ Purpose & Capability

The skill's name, README, SKILL.md and scripts consistently implement AIMLAPI chat + reasoning calls and require only AIMLAPI_API_KEY. However, the package metadata is incomplete: source is 'unknown', homepage is missing, and the embedded _meta.json ownerId/version differ from the registry metadata — this reduces provenance/traceability but does not change the script's stated purpose.

✓ Instruction Scope

SKILL.md and run_chat.py stay within scope: they build a /chat/completions payload, send it to api.aimlapi.com, and optionally write the response to a user-specified file. The script only reads AIMLAPI_API_KEY (env) or an --apikey-file if explicitly provided; it does not scan other system files or attempt to exfiltrate unrelated data.

✓ Install Mechanism

There is no install spec (instruction-only with one helper script). Nothing is downloaded or executed automatically — lowest-risk install footprint.

ℹ Credentials

The only required credential is AIMLAPI_API_KEY (declared as primaryEnv), which is proportionate. One caveat: the optional --apikey-file flag will read any file path you supply, so users must avoid pointing it at unrelated secret/config files.

✓ Persistence & Privilege

The skill does not request always:true, does not modify other skills or system configs, and does not persist credentials itself. It runs only when invoked.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install aiml-llm-reasoning
安装完成后，直接呼叫该 Skill 的名称或使用 /aiml-llm-reasoning 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.1

- No user-facing changes or file modifications in this version. - No updates to documentation, features, or workflows detected.

v1.0.0

Generate images or videos via AIMLAPI from prompts. Use when Codex needs reliable AI/ML API media generation with retries, explicit User-Agent headers, and async video polling.

元数据

Slug aiml-llm-reasoning

版本 1.0.1

许可证 —

累计安装 3

当前安装数 2

历史版本数 2

常见问题