Ask-More
v0.1.1Multi-LLM consultation for complex questions. Trigger when user sends "/ask [question]", "/ask" (no args, consult on recent conversation), "/ask --preset NAM...
⭐ 0· 85·0 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
The skill's name and description (multi-LLM consultation and synthesis) align with the provided config, prompt templates, and runtime steps. Required resources are limited to a models list (configured in the gateway) and optional local config/log files. Nothing in the repo requests unrelated credentials or binaries.
Instruction Scope
SKILL.md stays focused on packing context, spawning subagents (sessions_spawn / sessions_yield), collecting replies, and merging them. It does instruct the agent to read/write skills/ask-more/config.yaml and (optionally) write logs to skills/ask-more/logs/runs.jsonl. It will access recent conversation turns (for 'full' mode) and gateway provider model configs to check availability — these are necessary for the feature but are privacy-sensitive operations. The doc explicitly warns about not sending sensitive info.
Install Mechanism
There is no install script that downloads external code. The included scripts are small local shell helpers (config validator and logger) and config examples; no network downloads, third-party package installs, or obscure extract-from-URL steps are present.
Credentials
The skill requests no environment variables, no external credentials, and the config lists only model IDs. That is proportional for a multi-model consultation skill. The only potential privacy concern is that user questions are sent to multiple external model providers configured in the gateway (expected behavior).
Persistence & Privilege
always:false (normal). The skill writes/updates its own config.yaml (e.g., setting privacy_acknowledged) and can append run logs to a local file when enabled. This is expected for the feature but means the skill will create/modify files in its own directory; review logs/runs.jsonl if you enable logging.
Assessment
This skill appears to do what it says: it will read and (if needed) create or update skills/ask-more/config.yaml, check the gateway for configured models, spawn subagents (gateway calls) to send your packed question to each configured model, and (optionally) append run metadata to skills/ask-more/logs/runs.jsonl. Before enabling: 1) Review config.yaml and ensure the listed model/provider IDs are the ones you intend to consult. 2) Do not send highly sensitive secrets or personal data through /ask — the skill will forward the question to multiple external providers. 3) If you don't want local logging, set enable_logging:false or keep logs directory inaccessible. 4) If you prefer not to have the skill auto-write privacy_acknowledged, create config.yaml manually and set privacy_acknowledged:true. 5) Confirm you trust the gateway providers listed (the skill uses them to make external model calls). Overall the behavior is coherent with its purpose; exercise normal caution about sending sensitive content to third-party models.Like a lobster shell, security has layers — review code before you run it.
latestvk979ww0452qcwktgtnxw83vhdn83khv6
License
MIT-0
Free to use, modify, and redistribute. No attribution required.