Agent skill
ralph-preflight
Check all MCP servers and services before workflow execution. MUST pass before any Ralph command proceeds.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/ralph-preflight
SKILL.md
/ralph.preflight
Validate all required services before workflow execution.
CRITICAL: Run Before Any Ralph Command
This command MUST pass before /ralph.spawn, /ralph.integrate, or /ralph.cleanup.
If pre-flight fails, the workflow is BLOCKED until issues are resolved.
Process
-
Check Linear MCP (REQUIRED)
# Attempt to list teams - if this fails, Linear is not authenticated mcp__plugin_linear_linear__list_teams({})If empty or error: BLOCK immediately
# Verify floe team access mcp__plugin_linear_linear__get_team({"query": "floe"})If no access: BLOCK - user must fix Linear authentication
-
Check Git (REQUIRED)
bashgit status --porcelainIf fails: BLOCK - not in a git repository
-
Check Cognee (OPTIONAL)
Check if
.cognee/config.yamlexists andCOGNEE_API_KEYis set.If unavailable: WARN - memories will buffer locally via
.ralph/memory-buffer/ -
Check Memory Buffer
bashls .ralph/memory-buffer/pending/*.json 2>/dev/null | wc -l ls .ralph/memory-buffer/failed/*.json 2>/dev/null | wc -lIf failed entries: WARN - review
.ralph/memory-buffer/failed/ -
Check Manifest
Verify
.ralph/manifest.jsonis valid JSON.
Output: PASS
==================================================
RALPH WIGGUM PRE-FLIGHT CHECK: PASS
==================================================
[OK] linear: Connected to Linear MCP
[OK] git: Git repository valid
[OK] cognee: Cognee configuration found
[OK] manifest: Manifest valid (v1.0.0)
[OK] memory_buffer: Buffer empty - ready for use
--------------------------------------------------
Status: Ready to proceed
Output: BLOCKED
==================================================
RALPH WIGGUM PRE-FLIGHT CHECK: BLOCKED
==================================================
[!!] linear: Linear MCP not authenticated
Action: Re-authenticate Linear in Claude Code settings
[OK] git: Git repository valid
[??] cognee: COGNEE_API_KEY not set - using local buffer
Action: Set COGNEE_API_KEY environment variable
--------------------------------------------------
Status: BLOCKED - Fix required issues before proceeding
Recovery:
1. Fix the blocked service(s) above
2. Run: /ralph.preflight to verify
3. Resume with: /ralph.resume
Linear Authentication Recovery
If Linear MCP authentication fails:
- Check MCP server status in Claude Code settings
- Re-authenticate if token expired
- Verify team access - must have access to "floe" team
- Run
/ralph.preflightagain to verify
Cognee Degraded Mode
When Cognee is unavailable:
- Memories are written to
.ralph/memory-buffer/pending/ - Format:
{timestamp}-{type}-{id}.json - Auto-sync when Cognee reconnects (see
/ralph.memory-sync) - No context is lost - all decisions recorded locally
Pre-Flight Script
For programmatic access:
python .ralph/scripts/preflight.py # Check all
python .ralph/scripts/preflight.py -s linear # Check Linear only
python .ralph/scripts/preflight.py --json # JSON output
Configuration
From .ralph/config.yaml:
resilience:
preflight_checks:
linear: required # BLOCK if unavailable
cognee: optional # WARN + continue with buffer
git: required # BLOCK if unavailable
Related Commands
/ralph.spawn- Start agents (requires preflight PASS)/ralph.resume- Resume from saved session/ralph.memory-sync- Sync buffered memories/ralph.memory-status- Check buffer status
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?