powernode-platform / worker
CLAUDE.md4 starsChanged 18 days agoMITView on GitHub
Sidekiq standalone worker for Powernode. Query MCP before non-trivial worker changes (full protocol: conventions/mcp-first-workflow.md), using domain-specific queries: Query MCP first. Use these files when MCP returns no relevant results: Full tool catalog with parameters: reference/auto/mcp-tools.md. To load a tool's schema in a session, use ToolSearch (e.g. select:mcp__powernode__platform_search_knowledge).
The file
We split this CLAUDE.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
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
- Very specific: line 11
All AI execution jobs MUST include AiSuspensionCheckConcern for kill switch compliance (and call bail_if_ai_suspended!
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md 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
Worker CLAUDE.md
Sidekiq standalone worker for Powernode.
Critical Rules
- Jobs inherit from
BaseJob, implementexecute()method - API-only communication with server (mTLS for
/api/v1/internal/*; ActionCable WebSocket still uses JWT) Rails.logger- no puts/print# frozen_string_literal: truepragma required- All AI execution jobs MUST include
AiSuspensionCheckConcernfor kill switch compliance (and callbail_if_ai_suspended!first inexecute(); enforced byscripts/pattern-validation.shfor any executor — recallguidance-kill-switch-compliance)
MCP-First Worker Workflow
Query MCP before non-trivial worker changes (full protocol: conventions/mcp-first-workflow.md), using domain-specific queries:
Before Creating/Modifying
| Task | MCP Query |
| New job class | platform.discover_skills + platform.search_knowledge query: "background job patterns" |
| MCP-related jobs | platform.search_knowledge query: "MCP job" |
| Billing/payment jobs | platform.search_knowledge query: "billing jobs" |
| Job error handling | platform.query_learnings query: "Sidekiq error handling" — known failure modes |
| Job scheduling | platform.query_learnings query: "job scheduling" — cron patterns, race conditions |
| Autonomy maintenance jobs | platform.search_knowledge query: "autonomy jobs" — escalation timeout, goal maintenance, observation pipeline |
During Work
- Before new job patterns:
platform.query_learnings— check for established patterns and known pitfalls (retry storms, deadlocks, API timeouts) - Before job chain dependencies:
platform.search_knowledge_graph— understand existing job orchestration flows - Before API calls to server:
platform.search_knowledgequery: "worker API communication" — verify endpoint contracts
After Work (MANDATORY for non-trivial changes)
| Change type | Contribution |
| New job pattern | platform.create_learning category: pattern — document the approach |
| Job failure fix | platform.create_learning category: failure_mode — root cause + fix |
| Job chain/workflow | platform.extract_to_knowledge_graph — job dependencies, trigger conditions |
| Reusable job utility | platform.create_skill — codify the approach |
Context-Aware Documentation (file fallback)
Query MCP first. Use these files when MCP returns no relevant results:
| When working on | MCP Query | File Fallback |
app/jobs/* | platform.search_knowledge query: "background jobs" | guides/backend.md |
| MCP jobs | platform.search_knowledge query: "MCP tools" | concepts/mcp-and-tools.md |
MCP Tool Reference
Full tool catalog with parameters: reference/auto/mcp-tools.md. To load a tool's schema in a session, use ToolSearch (e.g. select:mcp__powernode__platform_search_knowledge).
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.

