src/hestai_mcp/_bundled_hub/library/skills/v9-agent-creation/SKILL.md
===SKILL:V9_AGENT_CREATION=== META: TYPE::SKILL VERSION::"1.0.0" STATUS::ACTIVE PURPOSE::"V9 blank-slate agent file creation per dream-team-architecture.md §2.2" SPEC_REFERENCE::docs/dream-team-architecture.md §1::CORE AUTHORITY::BLOCKING[v9_agent_file_commits⊕blank_slate_violations⊕phantom_task_profile_references<when_matrix_exists>] SCOPE::"Create and modify V9 agent .oct.md files per dream-team-architecture §2.2 blank-slate schema" // V9 agent files are ~50 line identity-only docum
npx skillsauth add elevanaltd/hestai-mcp src/hestai_mcp/_bundled_hub/library/skills/v9-agent-creationInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
===SKILL:V9_AGENT_CREATION=== META: TYPE::SKILL VERSION::"1.0.0" STATUS::ACTIVE PURPOSE::"V9 blank-slate agent file creation per dream-team-architecture.md §2.2" SPEC_REFERENCE::docs/dream-team-architecture.md
§1::CORE AUTHORITY::BLOCKING[v9_agent_file_commits⊕blank_slate_violations⊕phantom_task_profile_references<when_matrix_exists>] SCOPE::"Create and modify V9 agent .oct.md files per dream-team-architecture §2.2 blank-slate schema" // V9 agent files are ~50 line identity-only documents // Skills, patterns, and archetypes resolve dynamically via archetype-matrix config
§2::PROTOCOL
CREATION_SEQUENCE::[ 1::read_dream-team-architecture_§2.2[V9_schema_spec], 2::read_interview_assessment[from_agent-interview_skill_or_structured_equivalent], // agent-interview skill uses v8.1 terminology but its behavioral questions are schema-agnostic. // Interpret assessment findings through V9 lens (profiles not capabilities, identity not chassis). 3::select_cognition[LOGOS∨ETHOS∨PATHOS]→read_cognition_master_file, 4::author_v9_agent_file[§1::IDENTITY→§2::OPERATIONAL_BEHAVIOR→§3::TASK_PROFILES→§4::GRAMMAR], 5::author_archetype_matrix_entry[map_PROFILES_to_json_node⊕assign_archetypes_per_triad_physics_N1_or_N3⊕assign_skill_keys⊕assign_pattern_keys], 6::enforce_blank_slate[no_archetypes⊕no_skills⊕no_patterns_in_agent_file], 7::request_subject_agent_signoff, 8::write_via_octave_write ]
V9_STRUCTURE::[ META::[TYPE::AGENT_DEFINITION,VERSION,PURPOSE,CONTRACT], "§1::IDENTITY"::[ROLE,COGNITION,MISSION,PRINCIPLES[],AUTHORITY{ULTIMATE,BLOCKING,MANDATE}], "§2::OPERATIONAL_BEHAVIOR"::[TONE,MUST_ALWAYS[],MUST_NEVER[],DELIVERABLES[],EVIDENCE[],GATES[],HANDOFF,ESCALATION], "§3::TASK_PROFILES"::[PROFILES::[],DEFAULT::name], "§4::GRAMMAR"::[MUST_USE::[REGEX],MUST_NOT::[PATTERN]] ] // NO ARCHETYPE in §1 (blank slate — lives in archetype-matrix config) // NO MODEL_TIER (deployment config, not identity) // NO §3::CAPABILITIES (V9 uses §3::TASK_PROFILES) // NO §4::INTERACTION_RULES (V9 uses §4::GRAMMAR) // §2 is flat (no CONDUCT wrapper)
BLANK_SLATE_RULES::[ NO_ARCHETYPES::"§1::IDENTITY must NOT contain ARCHETYPE field — archetypes resolve via archetype-matrix", NO_SKILLS::"Agent file must NOT contain skill references — skills resolve via archetype-matrix profiles", NO_PATTERNS::"Agent file must NOT contain pattern references — patterns resolve via archetype-matrix profiles", NO_MODEL_TIER::"Agent file must NOT contain MODEL_TIER — deployment config is external", PROFILE_NAMES_ONLY::"§3::TASK_PROFILES contains profile NAME list + DEFAULT — no capability bindings" ]
TASK_PROFILE_RULES::[ PROFILES_is_array_of_names[string_identifiers_not_objects], DEFAULT_must_exist_in_PROFILES_array, each_name_should_map_to_archetype-matrix_entry<when_matrix_exists>, names_use_snake_case[code_writing∨test_building∨error_diagnosis] ]
ARCHETYPE_MATRIX_EDITING::[ PROTOCOL::[ 1::read_archetype-matrix_json[resources/starter-library/config/archetype-matrix.json], 2::identify_agent_role_key[snake-case_matching_agent_file_name], 3::create_profiles_object_with_one_entry_per_TASK_PROFILE, 4::per_profile_assign[archetypes⊕skills⊕patterns], 5::write_modified_json_preserving_existing_entries[Edit_tool_for_json], 6::validate_json_syntax[python3_json_load_check] ], JSON_SCHEMA::"matrix → agent-role → profiles → { profile_name: { archetypes: [], skills: [], patterns: [] } }", DATA_ENTRY_RULE::"Read file → Edit specific node → Write back. NEVER rewrite entire file. Preserve all other agent entries.", TRIAD_PHYSICS::[ "N=1::Single archetype with behavioral emphasis for tactical focus profiles (code_writing, gating, security_review)", "N=3::Triad of archetypes with behavioral emphasis for multi-objective profiles (orchestration, refactoring, strategic_planning)", "N=2::FORBIDDEN — binary deadlock creates unresolvable tension without synthesis axis", "FORMAT::Each archetype entry is a string with behavioral emphasis: ARCHETYPE<behavioral_application>" ], ARCHETYPE_FORMAT_EXAMPLES::[ "N=1_EXAMPLE::"HEPHAESTUS<implementation_craft>"", "N=3_EXAMPLE::["ODYSSEUS<cross_boundary_navigation>", "ATLAS<ultimate_accountability>", "APOLLO<system_foresight>"]" ] ]
AUTHORITY_STRUCTURE::[ ULTIMATE::"Capabilities this agent owns exclusively (list of domains)", BLOCKING::"Conditions that MUST be blocked (list of violations)", MANDATE::"Single sentence: what this agent prevents from happening" ]
COGNITION_RULES::[ single_cognition_only[LOGOS∨ETHOS∨PATHOS], cognition_master_provides[FORCE⊕ESSENCE⊕ELEMENT⊕MODE⊕PRIME_DIRECTIVE⊕CRAFT⊕THINK⊕THINK_NEVER], agent_files_do_NOT_duplicate_cognition_properties, COGNITION_field_in_§1_is_link_key_to_library_cognitions_TYPE_oct_md ]
SIZE_TARGET::"~50 lines — lean identity only. Everything that resolves dynamically is excluded."
§3::GOVERNANCE
NEVER::[ author_without_interview_assessment, include_ARCHETYPE_in_§1[blank_slate_violation], include_skills_or_patterns_in_agent_file[blank_slate_violation], include_MODEL_TIER[deployment_config_not_identity], use_§3::CAPABILITIES[V9_uses_§3::TASK_PROFILES], use_§4::INTERACTION_RULES[V9_uses_§4::GRAMMAR], include_CONDUCT_wrapper_in_§2[V9_§2_is_flat], phantom_task_profile_references<when_matrix_exists>, Write∨Edit_for_oct_md[use_octave_write], exceed_50_line_target_without_justification ]
VALIDATION::[ V9_structure_compliance[all_4_sections_present], blank_slate_verified[no_archetypes⊕no_skills⊕no_patterns], DEFAULT_profile_exists_in_PROFILES_array, AUTHORITY_has_ULTIMATE⊕BLOCKING⊕MANDATE, §2_is_flat[no_CONDUCT_wrapper], cognition_is_link_key_only[no_duplicated_properties], file_approximately_50_lines[~50_target_with_justification_for_exceptions] ]
§5::ANCHOR_KERNEL TARGET::v9_blank_slate_agent_file_creation NEVER::[archetypes_in_agent_file,skills_in_agent_file,patterns_in_agent_file,MODEL_TIER_in_agent_file,§3::CAPABILITIES_structure,§4::INTERACTION_RULES_structure,CONDUCT_wrapper,phantom_task_profile_references<when_matrix_exists>,authoring_without_assessment,Write∨Edit_for_oct_md] MUST::[read_dream-team-architecture_first,interview_evidence,enforce_blank_slate,author_matrix_entry[archetypes_N1_or_N3⊕skills⊕patterns],subject_agent_signoff,~50_line_target,flat_§2_structure,AUTHORITY{ULTIMATE⊕BLOCKING⊕MANDATE},octave_write_for_oct_md] GATE::"V9 compliant blank-slate agent file with §1::IDENTITY→§2::OPERATIONAL_BEHAVIOR→§3::TASK_PROFILES→§4::GRAMMAR, archetype-matrix entry authored with triad physics, subject agent signoff obtained?"
===END===
tools
Supabase test harness patterns including local Supabase setup, test user creation via Auth Admin API, RLS testing, migration testing (db_reset), seed sequences, rate limiting, and environment detection. Use when setting up Supabase testing infrastructure, creating test users, troubleshooting Supabase test failures, or implementing RLS validation. Triggers on: supabase test setup, test user creation, supabase local testing, RLS testing, migration testing, supabase test harness, auth test helpers.
tools
Supabase operational knowledge for migrations, RLS optimization, MCP tool benchmarks, and ADR-003 compliance. Use when validating database migrations, optimizing Row-Level Security policies, checking MCP tool performance, or ensuring Supabase operational standards. Triggers on: migration validation, RLS patterns, Supabase benchmarks, ADR-003, database state tracking, schema governance.
tools
Vercel preview deployment access patterns including automation bypass, protected preview authentication, and CI integration
tools
===SKILL:VALIDATION_METHODOLOGY=== META: TYPE::SKILL VERSION::"1.0" PURPOSE::"Rigorous 6-step reality enforcement protocol" §1::CORE_PROTOCOL VALIDATION_METHODOLOGY::[ STEP_1_NATURAL_LAW::"Identify physics constraints, thermodynamic limits, information theory bounds", STEP_2_RESOURCE_REALITY::"Inventory actual available resources (time, people, tools, budget) with evidence", STEP_3_CAPABILITY_ASSESSMENT::"Validate team skills, tool availability, knowledge gaps against requirements",