aaa1dbf3d0
Schema v2 bump (per sot/04-STANDARDS/DISCLOSURE-SCHEMA.md §4.6) adds the external_orchestrators surface. Sandcastle was previously parked in DISCLOSURE.md §12.1 "Pending JP review"; Wave-7 Q2 resolved the open question in favor of (b) schema §4.6's dedicated external_orchestrators taxonomy (cleaner separation from HTTP/gRPC sovereign_apis). Changes: - manifest.yaml: disclosure.schema_version 1 → 2; add external_orchestrators with sandcastle entry (transport=cli, mode=exec, version_pin=v0.5.11, sandboxed=true, hosted_api=anthropic, called_by lib/cto-worker.sh). - DISCLOSURE.md: new canonical §6.5 External orchestrators (sandcastle row + governance/pin/check-6.e notes); §5 footer note updated (no longer pending); §9 drift table adds external_orchestrators row; §12.1 marked RESOLVED with audit-trail stub; last_reviewed bumped to 2026-05-25. Pin v0.5.11 matches external_tool_deps[0].pin and the workspace CLAUDE.md hard rule (sandcastle read-only; bumps human-only). sot-precommit clean. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
284 lines
12 KiB
YAML
284 lines
12 KiB
YAML
# CTO profile distribution manifest — machine-readable identity + install contract.
|
|
# Read by install.sh. Convention shared by all Hermes profile distributions
|
|
# (see ../sot/03-PROTOCOLS/PROFILE-DISTRIBUTION-PROTOCOL.md — the canonical protocol).
|
|
profile: cto-planb # Hermes profile name (org-scoped); see also distribution.yaml → name
|
|
kind: profile-distribution # family marker; CTO = third C-suite profile (after CMO + CEO)
|
|
role: cto # function; same skill bundle could deploy as cto-<other-org>
|
|
org: planb # org scope — this profile serves Plan B
|
|
version: 1.0.0 # MVP — executable cto-agent skill + cto-worker.sh helper + 2 toolkit skills
|
|
identity: AGENT.md # WHO (role, mission, boundaries)
|
|
contract: CONTRACT.md # behavior contract — tier T1 (this file wins)
|
|
|
|
# Governance — org / owner / vision / rules linked to SOT (PROFILE-DISTRIBUTION-PROTOCOL §2.2).
|
|
governance:
|
|
org: planb
|
|
owner: jp
|
|
approval_authority: jp
|
|
vision_refs:
|
|
- ../sot/01-ROADMAP/CORTEX-OS-ROADMAP.md
|
|
- ../sot/02-FRAMEWORK/CORTEX-OS-FRAMEWORK.md
|
|
governing_protocols:
|
|
- ../sot/03-PROTOCOLS/PROFILE-DISTRIBUTION-PROTOCOL.md
|
|
standards:
|
|
- ../sot/04-STANDARDS/FRONTMATTER-SPEC.md
|
|
- ../sot/04-STANDARDS/SOT-ENFORCEMENT.md
|
|
brand_master_ref: ../sot/07-BRAND/PLANB-BRAND-SYNTHESIS.md
|
|
north_star: "reliable, evolving tech — sandcastle-orchestrated code work, JP-approved deploys, never bypass isolation"
|
|
|
|
skills: # exposed to Hermes via skills.external_dirs (→ <repo>/skills)
|
|
- skills/cto-agent # orchestrator (loop operator)
|
|
- skills/cto-python-toolkit # Python stack patterns (closes Python gap — inline until cortex/ lib extracted)
|
|
- skills/cto-angular-toolkit # Angular stack patterns (closes Angular gap — anchored to adwright-console)
|
|
|
|
# Role tools = scripts at repo root (the "lib"), reached through credbridge.
|
|
lib:
|
|
- credbridge.sh # credctl → env → CLI (secrets never logged)
|
|
|
|
# External read-only siblings + cortex/ tooling consumed by this profile.
|
|
# Stacks: typescript (sandcastle), dotnet (CQRS), dart (Flutter/gRPC), go (libs+QA), rust (runtime), multi (gates/bash/cortex).
|
|
# Python + Angular have no specific cortex/ tooling yet — CTO handles them via sandcastle generic Claude Code path.
|
|
external_tool_deps:
|
|
# Agent orchestration (external — Matt Pocock, MIT)
|
|
- repo: sandcastle
|
|
path: ../sandcastle
|
|
stack: typescript
|
|
pin: v0.5.11
|
|
role: sandboxed agent orchestrator for code-modifying work (Docker/Podman/Vercel)
|
|
registry: ../sot/06-REGISTRY/EXTERNAL-REFS/SANDCASTLE.md
|
|
|
|
# .NET / C# (Plan B's primary backend stack)
|
|
- repo: L6-svrnty.lib-dotnet-cqrs
|
|
path: ../../cortex/L6-svrnty.lib-dotnet-cqrs
|
|
stack: dotnet
|
|
role: CQRS framework (.NET 10, 21 projects, gRPC source-gen, FluentValidation, Minimal API)
|
|
- repo: L5-svrnty.tool-cqrs-plugin
|
|
path: ../../cortex/L5-svrnty.tool-cqrs-plugin
|
|
stack: dotnet
|
|
role: Claude Code plugin for CQRS/gRPC scaffolding + validation patterns
|
|
- repo: pi-bte-plugin
|
|
path: ../../cortex/pi-bte-plugin
|
|
stack: dotnet+design
|
|
role: Pi plugin — DTCG validation, voice schema lint, DESIGN.md export, contrast check, .NET 10 build verification
|
|
|
|
# Dart / Flutter (mobile + desktop client stack)
|
|
- repo: L6-svrnty.lib-cqrs-datasource
|
|
path: ../../cortex/L6-svrnty.lib-cqrs-datasource
|
|
stack: dart
|
|
pin: v1.0.1
|
|
role: gRPC client bridging Flutter apps to .NET CQRS backends (auto proto gen)
|
|
|
|
# Go (sovereign core stack — runtime, creds, memory, QA)
|
|
- repo: L6-svrnty.lib-llm
|
|
path: ../../cortex/L6-svrnty.lib-llm
|
|
stack: go
|
|
role: multi-provider LLM interface (Ollama/Claude/OpenAI/Gemini) + A/B harness
|
|
- repo: L6-svrnty.core-credentials
|
|
path: ../../cortex/L6-svrnty.core-credentials
|
|
stack: go
|
|
role: encrypted credential store (AES-256-GCM) — credctl backend
|
|
- repo: L6-svrnty.core-memory
|
|
path: ../../cortex/L6-svrnty.core-memory
|
|
stack: go
|
|
role: long-term memory + semantic search (ONNX embeddings) for agent fleet
|
|
- repo: PG-svrnty.tool-qa
|
|
path: ../../cortex/PG-svrnty.tool-qa
|
|
stack: go
|
|
role: QA orchestrator — YAML-defined test scenarios for zeroclaw/voice/gRPC services
|
|
|
|
# Rust (zero-overhead runtime layer)
|
|
- repo: L6-svrnty.core-runtime
|
|
path: ../../cortex/L6-svrnty.core-runtime
|
|
stack: rust
|
|
pin: v0.1.7
|
|
role: agent runtime (5MB RAM target, Rust/Tokio) — zeroclaw CLI
|
|
|
|
# Multi-stack utility
|
|
- repo: PG-svrnty.lib-quality-gates
|
|
path: ../../cortex/PG-svrnty.lib-quality-gates
|
|
stack: multi
|
|
role: 48 portable quality gates across Go/Rust/Dart/Python/C#/Docker/Proto (Bash 4+)
|
|
- repo: L5-svrnty.lib-skills-engineering
|
|
path: ../../cortex/L5-svrnty.lib-skills-engineering
|
|
stack: multi
|
|
role: 28 engineering patterns — CQRS, gRPC, Flutter design, saga, events, error handling
|
|
- repo: L5-svrnty.tool-bash-plugin
|
|
path: ../../cortex/L5-svrnty.tool-bash-plugin
|
|
stack: bash
|
|
role: Bash script engineering plugin (9 categories — init/gate/hook/cron/probe/seal/deploy/test/orchestrate)
|
|
# PC-svrnty.tool-cortex-plugin REMOVED 2026-05-24 (Wave-4 orphan cleanup) — never cited in any cto skill body
|
|
# See sot/06-REGISTRY/audits/RECOMMENDATIONS-cto-2026-05-24.md §0.2 + §4 C13
|
|
|
|
# Stacks NOT yet covered by dedicated cortex/ tooling:
|
|
# - Python: handled via sandcastle generic Claude Code path; no Python framework lib
|
|
# - Angular: handled via sandcastle generic Claude Code path; no Angular framework lib
|
|
# CTO declares these gaps in CONTRACT.md §6 (Tech stacks supported).
|
|
|
|
requires_tools: [terminal, memory_tool]
|
|
|
|
db:
|
|
file: cto.db # runtime state; created from schema.sql; never committed
|
|
schema: schema.sql
|
|
|
|
cron: [] # no cron at launch — CTO runs on-demand or on kanban tick
|
|
|
|
config: # portable per-install settings
|
|
- key: github_org
|
|
prompt: GitHub org for code work (e.g., svrnty)
|
|
default: "svrnty"
|
|
- key: default_sandcastle_provider
|
|
prompt: Default sandcastle sandbox provider (docker|podman|vercel|noSandbox)
|
|
default: "docker"
|
|
|
|
credentials: # provisioned via `credctl set <name>` — never shipped
|
|
required: [] # v1 has no required creds (no deploy/cloud yet)
|
|
optional: [github-pat] # for opening PRs via gh CLI when CTO ships v2
|
|
|
|
# Disclosure block (Wave-4 — per sot/04-STANDARDS/DISCLOSURE-SCHEMA.md schema_version 1).
|
|
# Authoritative runtime-truth contract; pre-push hook check 6 verifies declared == live.
|
|
# Derived from Wave-3 recommendations: sot/06-REGISTRY/audits/RECOMMENDATIONS-cto-2026-05-24.md
|
|
disclosure:
|
|
scope: org
|
|
schema_version: 2 # bumped Wave-7 D2 (2026-05-25) — adds external_orchestrators surface per DISCLOSURE-SCHEMA §4.6
|
|
chat_facing: false # cto is kanban-driven; JP chats with steev, not cto (CONTRACT.md §3)
|
|
delegates_to: [] # cto consumes sandcastle as a tool, not a sub-agent (CONTRACT.md §1, §9)
|
|
inherit_builtins: false # deny-by-default; cto has zero builtins enabled
|
|
inherit_mcp_toolsets: false # deny-by-default; closes the bte-MCP-leak risk seen on ceo/steev
|
|
sovereign_only: false # INTENTIONAL — cto uses claudeCode('claude-opus-4-7') INSIDE sandcastle
|
|
# isolation (CONTRACT.md §5). cto-agent itself runs sovereign qwen3.6.
|
|
inherit_dirs: [] # no external_dirs
|
|
|
|
skills:
|
|
- id: cto-agent
|
|
source: local
|
|
path: skills/cto-agent
|
|
role: orchestrator
|
|
- id: cto-python-toolkit
|
|
source: local
|
|
path: skills/cto-python-toolkit
|
|
role: toolkit
|
|
justification: "Python stack patterns — closes CONTRACT.md §6 'Python = skill-only' gap; anchored to bte-mcp, svrnty-hermes-webui-plugin, curator/sweep.py, scripts/sot-precommit.py"
|
|
- id: cto-angular-toolkit
|
|
source: local
|
|
path: skills/cto-angular-toolkit
|
|
role: toolkit
|
|
justification: "Angular stack patterns — closes CONTRACT.md §6 'Angular = skill-only' gap; anchored to adwright/adwright-console"
|
|
|
|
mcp_servers: [] # cto orchestrates via sandcastle + shell, not MCP
|
|
|
|
sovereign_apis:
|
|
- name: bte-rest
|
|
endpoint: "http://localhost:5000"
|
|
transport: http
|
|
mode: read-write
|
|
called_by:
|
|
- skills/cto-agent/SKILL.md
|
|
- skills/cto-angular-toolkit/SKILL.md
|
|
justification: "BTE REST endpoint /api/export-design-md — cited as the DESIGN.md emit path for UI tasks; not auto-invoked at v1.0 (documented pattern only)"
|
|
|
|
cortex_tools:
|
|
- id: L6-svrnty.lib-dotnet-cqrs
|
|
stack: dotnet
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: ".NET CQRS routing target — sandcastle sub-agent reads patterns when mounted"
|
|
- id: L5-svrnty.tool-cqrs-plugin
|
|
stack: dotnet
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: ".NET scaffolding plugin — routing target"
|
|
- id: pi-bte-plugin
|
|
stack: dotnet
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
- skills/cto-angular-toolkit/SKILL.md
|
|
justification: "DTCG validation + voice schema lint + DESIGN.md export — routing target + DESIGN.md emit path"
|
|
- id: L6-svrnty.lib-cqrs-datasource
|
|
stack: dart
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
- skills/cto-angular-toolkit/SKILL.md
|
|
justification: "Flutter gRPC client + Angular gRPC-web reference — routing target"
|
|
- id: L6-svrnty.lib-llm
|
|
stack: go
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: "Go multi-provider LLM interface — routing target for Go tasks"
|
|
- id: L6-svrnty.core-credentials
|
|
stack: go
|
|
invoked_at_runtime: true
|
|
mode: read+exec
|
|
referenced_in:
|
|
- credbridge.sh
|
|
justification: "Runtime-invoked via credctl CLI from credbridge.sh — every cmd_open_pr resolves github-pat through this lib"
|
|
- id: L6-svrnty.core-memory
|
|
stack: go
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: "Go memory lib — routing target; requires_tools memory_tool is Hermes-side, not direct call"
|
|
- id: PG-svrnty.tool-qa
|
|
stack: go
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: "QA orchestrator — routing target for Go QA work"
|
|
- id: L6-svrnty.core-runtime
|
|
stack: rust
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: "zeroclaw runtime — routing target for Rust tasks"
|
|
- id: PG-svrnty.lib-quality-gates
|
|
stack: multi
|
|
invoked_at_runtime: true
|
|
mode: read+exec
|
|
referenced_in:
|
|
- skills/cto-python-toolkit/SKILL.md
|
|
- skills/cto-angular-toolkit/SKILL.md
|
|
justification: "Runtime-invoked post-sandcastle via $QG/bin/run-gates --stack python|typescript --repo X --branch Y"
|
|
- id: L5-svrnty.lib-skills-engineering
|
|
stack: multi
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: "28-pattern engineering reference — routing target"
|
|
- id: L5-svrnty.tool-bash-plugin
|
|
stack: bash
|
|
invoked_at_runtime: false
|
|
mode: read
|
|
referenced_in:
|
|
- skills/cto-agent/SKILL.md
|
|
justification: "Bash scripting plugin — routing target for Bash tasks"
|
|
|
|
credentials: [] # github-pat declaration parked under Pending JP review in DISCLOSURE.md §12
|
|
# (cred-adjacent PAUSE per Wave-3 recommendations §5 K1)
|
|
|
|
# External orchestrators (schema v2+ — Wave-7 D2). Sandcastle is cto's primary
|
|
# execution mechanism (CONTRACT.md §5). sandboxed=true + sovereign_only=false
|
|
# = the 4-layer safety stack (sandbox isolation + git branch + PR + JP approval).
|
|
external_orchestrators:
|
|
- id: sandcastle
|
|
transport: cli
|
|
mode: exec
|
|
called_by:
|
|
- lib/cto-worker.sh
|
|
version_pin: v0.5.11
|
|
sandboxed: true
|
|
sovereign_required: false
|
|
hosted_api: anthropic
|
|
justification: "isolated claudeCode exec per CONTRACT.md §5 (escape valve under sovereign_only=false)"
|