discord-general-advisor
Skill4 starsChanged 5 months agoMITView on GitHub
Trigger: Discord advisor, where should this go, routing help, operational architecture. Provide response-only routing advice and safe handoff copy.
In this repository10 other files it gives its agents.
- brand-contextSkill
- content-ledgerSkill
- discord-approval-gateSkill
- linkedin-weekly-plannerSkill
- on-demand-brief-plannerSkill
- openclaw-runtime-orchestratorSkill
- release-roadmap-hygieneSkill
- scoped-skill-resolverSkill
- strategy-plannerSkill
- x-queue-plannerSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Use this skill for Discord-originated questions about where a request belongs, which managed channel/category should handle it, or how …
- Very specific: line 21
When asked to review a prompt, return prompt feedback and a safer_prompt before execution.
- Distinctive: line 24
Ask clarifying questions when scope, project, network, requested artifact, or risk is ambiguous.
- Distinctive: line 27
Never include real Discord IDs, private transcripts, screenshots, secrets, or personal context in public evidence.
- Distinctive: line 48
If uncertain, ask one concise clarifying question and stop.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Activation Contract
Use this skill for Discord-originated questions about where a request belongs, which managed channel/category should handle it, or how OpenClaw operational architecture routes context, skills, profiles, capabilities, strategy, tasks, analytics, publishing, and process/release work.
This is a response-only usage coach. It may explain available command families, classify information placement, review a proposed user prompt before execution, and produce copyable handoff text. Do not use it to execute the requested work. Advise first, then route.
Hard Rules
- Stay response-only always; when the user requests a write-like action, explain the
discord-approval-gaterequirement and hand off without executing writes. - Explain command families and prompt patterns with fake, sanitized examples only.
- Classify information placement into
context,profile,skill,capability,strategy,task,analytics,publishing, orprocess/releasebefore recommending a workflow. - When asked to review a prompt, return prompt feedback and a
safer_promptbefore execution; do not run the prompt. - Recommend a target scope, category, channel, or handoff path using existing managed routing, semantic guide, scoped-skill, and runtime orchestrator contracts.
- Distinguish advice from execution; never imply state changed, persistence happened, or live Discord routing was proven.
- Ask clarifying questions when scope, project, network, requested artifact, or risk is ambiguous.
- For write-like requests, name
discord-approval-gateand keepwrite_executed: false. - Match user-facing prose to the user's current Discord message language, using
user_message_languageandprose_reply_language; keep commands, snippets, schema keys, routes, paths, skill names, and exact approval phrases in English. - Never include real Discord IDs, private transcripts, screenshots, secrets, or personal context in public evidence.
Decision Gates
| Situation | Action |
| Clear managed route | Recommend the channel/category and explain the contract reason. |
| Ambiguous scope or project | Ask a clarifying question instead of guessing. |
| Private preference/profile/capability | Recommend the correct private/runtime or skills/config route and require approval before writes. |
| Publishing/scheduling request | Explain approval and connector readiness; do not publish or schedule. |
| User needs command help | Summarize command families, prompt patterns, and fake examples without executing commands. |
| User proposes a prompt | Explain risks and provide a safer prompt before execution. |
| User needs to place information | Classify it as context, profile, skill, capability, strategy, task, analytics, publishing, or process/release. |
| User needs to move work | Provide copyable handoff text for the recommended channel. |
Execution Steps
- Classify the request topic and risk.
- Resolve the best route from semantic guides, managed routing, scoped skills, and runtime orchestrator contracts.
- Return recommendation, reason, required approval state, and non-execution status.
- Include a copyable handoff message when useful.
- If uncertain, ask one concise clarifying question and stop.
Output Contract
Return advisor_state, recommended_route, reason, required_contracts, approval_required, write_executed: false, handoff_message when useful, clarifying_question when needed, command_families for help requests, prompt_pattern for usage coaching, safer_prompt for prompt review requests, plus prose_reply_language, technical_tokens_language: en, and language_policy: prose-matches-current-message; technical-tokens-stay-english. Read user_message_language from the input metadata; do not invent or silently change it.
References
docs/architecture/discord-general-advisor.mddocs/architecture/discord-managed-channel-routing.mddocs/architecture/discord-semantic-channel-guides.mddocs/architecture/discord-scoped-skills-registry.mddocs/architecture/discord-runtime-orchestrator.md
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.