openclaw-config-advisor
The openclaw-config-advisor skill diagnoses configuration issues in OpenClaw deployments by performing read-only scans of provider routing, fallback models, channels, MCP servers, and gateway settings. Use this skill when OpenClaw exhibits inconsistent behavior, missing tools, connection failures, or unexpected model routing before making configuration changes.
git clone --depth 1 https://github.com/ArchieIndian/openclaw-superpowers /tmp/openclaw-config-advisor && cp -r /tmp/openclaw-config-advisor/skills/openclaw-native/openclaw-config-advisor ~/.claude/skills/openclaw-config-advisorSKILL.md
# OpenClaw Config Advisor Config problems usually look like model failures, missing tools, or broken channels. Use this skill when OpenClaw behaves inconsistently and you need a read-only diagnosis before changing settings. ## When to Invoke - Provider routing or fallback model behavior is wrong - A channel cannot connect or stops receiving messages - MCP tools are missing, stale, or unavailable - The gateway starts but behaves differently than expected - You need a current config summary before editing OpenClaw settings ## Checks | Area | What to inspect | |---|---| | Providers | Provider keys, model names, default model hints | | Fallback | Fallback or backup model settings | | Channels | Telegram, Discord, Slack, WhatsApp, Signal, or Lark entries | | MCP | MCP server declarations and command/url hints | | Gateway | Host, port, log, and process configuration hints | ## How to Use ```bash python3 advise.py --scan python3 advise.py --scan --config-dir ~/.openclaw python3 advise.py --explain fallback python3 advise.py --status python3 advise.py --format json ``` ## Procedure 1. Run `python3 advise.py --scan --format json`. 2. Review findings by severity: `WARN` first, then `INFO`. 3. Inspect the named config files before editing anything. 4. Apply config changes only after the user confirms the intended routing or channel behavior. 5. Re-run the scan and compare with the prior state summary. ## State Scan summaries are stored in `~/.openclaw/skill-state/openclaw-config-advisor/state.yaml`. Fields: `last_scan_at`, `config_dir`, `findings`, `affected_areas`, and `scan_history`. ## Guardrails - This skill is read-only; it does not modify OpenClaw config. - Missing config directories are warnings, not crashes. - Do not print secrets. Report that secret-like values exist without echoing them.
Syncs agent daily memory and MEMORY.md to an Obsidian vault so notes are human-browsable. Use nightly or on demand.
Structured ideation before any implementation. Use when starting any non-trivial task.
Scaffolds and validates new superpowers skills. Use when creating a new skill for this repository.
Executes plans task-by-task with verification. Use when implementing a plan.
Triggers a secondary verification pass for any agent output containing factual claims, numbers, dates, or named entities before the output is acted on
Crawls a new codebase to infer stack, conventions, and key invariants, then generates a PROJECT.md context file for the agent
Handles PR review feedback by fetching comments, grouping issues, fixing one group at a time, and verifying before replies.
Detects skill name shadowing and description-overlap conflicts that cause OpenClaw to trigger the wrong skill or silently ignore one when two skills compete for the same intent.