elevanaltd/src/hestai_mcp/_bundled_hub/library/skills/rccafp

===SKILL:RCCAFP=== META: TYPE::SKILL VERSION::"1.0" PURPOSE::"Root Cause, Corrective Action, and Future Proofing incident management protocol"

§1::PROTOCOL_OVERVIEW RCCAFP::[ FULL_NAME::"Root Cause, Corrective Action, and Future Proofing", ORIGIN::"Daedalus-era LLM Overreach incident (2025-05-26)", SCOPE::"Incident management for multi-agent AI systems with human orchestration oversight", PRINCIPLES::[ "Systematic evidence collection with exact causality chains", "Multi-agent constraint validation via ETHOS boundary enforcement", "LLM-optimized structure for efficient cross-reference", "Prevention-focused learning transforms incidents into architectural improvements", "Human authority preservation throughout process" ] ]

§2::FIVE_PHASE_PROCESS PHASES::[ P1_DETECTION_TRIAGE::[ ACTION::"Create incident directory YYYYMMDD-[incident-name]", STRUCTURE::"01-root-cause/ 02-corrective-action/ 03-future-proof/", OUTPUT::"Initial symptom documentation and severity assessment" ], P2_ROOT_CAUSE::[ ACTION::"Systematic causality chain analysis with evidence collection", VALIDATION::"Cross-validation through multi-agent assessment", OUTPUT::"Causality chain documentation with artifacts" ], P3_CORRECTIVE_ACTION::[ ACTION::"Immediate containment and mitigation", VALIDATION::"ETHOS constraint validation and gap identification", OUTPUT::"Process modification documentation and triage execution" ], P4_FUTURE_PROOFING::[ ACTION::"Architectural learning extraction and prevention mechanism design", VALIDATION::"System improvement evidence synthesis", OUTPUT::"Prevention mechanisms and monitoring requirements" ], P5_CLOSURE::[ ACTION::"Verification and synthesis gate processing", GATE::"Findings pass through synthesis gate before system integration", OUTPUT::"Organizational learning documentation and protocol refinement" ] ]

§3::INCIDENT_DIRECTORY TEMPLATE::[ ROOT::".hestai/state/incidents/YYYYMMDD-[incident-name]/", DIRS::[ "01-root-cause/", "02-corrective-action/", "03-future-proof/" ], REQUIRED_FILES::[ "01-root-cause/causality-chain.md", "02-corrective-action/constraint-validations.md", "02-corrective-action/triage-execution.md", "03-future-proof/prevention-mechanisms.md" ] ]

§4::SYNTHESIS_GATE GATE::[ PURPOSE::"Controlled integration of incident learnings into system documentation", FLOW::"findings → .hestai/state/incidents/pending/ → review → validated integration", REQUIREMENT::"All findings must pass synthesis gate before promotion to governance docs", ANTI_OVERREACH::[ "Fixed document structure prevents documentation expansion", "Evidence-based requirement limits speculative content", "Synthesis gate controls integration of learnings" ] ]

§5::AGENT_MAPPING ROLES::[ HOLISTIC_ORCHESTRATOR::"Orchestrates RCCAFP process, owns gap accountability", CRITICAL_ENGINEER::"Leads root cause investigation (tactical analysis)", PRINCIPAL_ENGINEER::"Leads future proofing (strategic prevention)", IMPLEMENTATION_LEAD::"Executes corrective action implementations", REQUIREMENTS_STEWARD::"Validates learnings against North Star before integration" ]

§6::SUCCESS_METRICS METRICS::[ MTTD::"Mean Time to Detection of system boundary violations", MTTC::"Mean Time to Containment for incidents", COMPLETENESS::"Documentation accuracy and completeness score", VALIDATION::"Constraint validation effectiveness", RECURRENCE::"Incident recurrence rate for similar failures", IMPROVEMENT::"Architecture improvement rate from incident analysis" ]

§7::ANTI_PATTERNS AVOID::[ SILENT_FIX::"Fixing incidents without documenting root cause", SKIP_SYNTHESIS::"Promoting learnings without synthesis gate review", BLAME_CASCADE::"Attributing failures to agents rather than system gaps", SPECULATIVE_PREVENTION::"Creating prevention mechanisms without evidence", SCOPE_CREEP::"Expanding incident response beyond the specific failure" ]

===END===