Foundation Step 0: UC Resources (Schemas + Volumes)
Cross-cutting. This skill is consumed by every Foundation step and
agent track. It exists so that no downstream skill (KA, agent memory, tool
exports, benchmark persistence, prompt registry) has to repeat schema or
volume provisioning. If you find yourself writing CREATE SCHEMA or
CREATE VOLUME inside another skill, delegate to F0 instead.
What This Skill Owns
| Resource |
Purpose |
Default name |
| Agent UC schema |
Holds agent-owned assets — KA source volume, tool output volume, benchmark tables, prompt registry pointers, memory tables |
${user_schema_prefix}_agent |
| Ops UC schema |
Holds operational assets — MLflow OTeL trace tables (F2), monitoring tables (SDLC 07), eval-run rollups (SDLC 04) |
${user_schema_prefix}_ops |
knowledge_sources MANAGED volume (in agent schema) |
Source markdown / PDFs for Knowledge Assistants (F5) and any retrieval-style tool |
/Volumes/${catalog}/${agent_schema}/knowledge_sources |
agent_outputs MANAGED volume (in agent schema) |
CSV exports, charts, generated artifacts produced by @function_tool calls |
/Volumes/${catalog}/${agent_schema}/agent_outputs |
signoffs MANAGED volume (in ops schema) |
Stakeholder + engineering sign-off decision.md artifacts (SDLC 04b) — the durable, UC-governed location read by SDLC 06 promotion gates |
/Volumes/${catalog}/${ops_schema}/signoffs |
| Custom volumes (in agent or ops schema) |
Anything declared in state://AgentSpec.required_volumes[] |
/Volumes/${catalog}/${schema}/${name} |
signoffs is generic. F0 pre-creates the volume under a workshop-neutral
name (signoffs) so SDLC 04b never has to provision its own UC location and
SDLC 06 can resolve /Volumes/${catalog}/${ops_schema}/signoffs/v<N>/decision.md
deterministically. The volume name does not carry any use-case-specific
prefix (no skyloyalty_signoffs, no ${use_case_slug}_signoffs); the
per-version directory v<N>/ carries the disambiguation.
Every operation is idempotent (CREATE SCHEMA IF NOT EXISTS,
CREATE VOLUME IF NOT EXISTS, both as DDL — an existing object is a no-op). It
is safe to call this skill from any prompt that needs to be sure these
resources exist; the cost on a warm workspace is one round-trip per resource.
What This Skill Does NOT Own
- Catalog creation. A UC catalog must already exist; F0 reads
state://Resources.uc.catalog and validates the caller has USE CATALOG +
CREATE SCHEMA on it.
- UC table DDL. Tables for OTeL traces, benchmarks, memory, or monitoring
are owned by their respective consuming skills (F2, SDLC 02/04/07).
- Source content. F0 creates an empty
knowledge_sources volume; F5 (or
any other consumer) is responsible for populating it with files.
- Bronze data. Bronze schema + Delta tables are owned by the bronze layer
setup skill (
data_product_accelerator/skills/bronze/00-bronze-layer-setup).
When to Use
- Always, near the top of the prompt sequence — before F1 (MLflow env) so
that anything downstream can assume schemas + volumes exist.
- Whenever you add a new consuming skill that needs a new volume — extend
state://AgentSpec.required_volumes[] rather than putting CREATE VOLUME
inside the consuming skill.
- After
bootstrap when the resolved spec is available, but before any other
Foundation step.
Prerequisites
| Requirement |
How to Check |
state://Resources.uc.catalog is resolved |
Set by vibecoding-state bootstrap op |
state://Resources.uc.user_schema_prefix is resolved |
Derived in bootstrap from the user identity + use_case_slug |
| Workshop SQL warehouse is RUNNING |
state://Resources.warehouse_id (set by Prompt 1 preflight) — needed for CREATE SCHEMA execution via WorkspaceClient.statement_execution |
databricks-sdk >= 0.30 installed |
python -c "from databricks.sdk import WorkspaceClient; from databricks.sdk.service.sql import StatementState" |
Caller has USE CATALOG + CREATE SCHEMA on the catalog |
databricks unity-catalog catalogs get $CATALOG --output json | jq '.privileges' |
Inputs
# Resolved by the caller from state — F0 does not read state directly.
uc_catalog: "string (required)" # e.g. "main"
user_schema_prefix: "string (required)" # e.g. "jane_d_booking_app"
agent_schema: "string (required)" # default: "${user_schema_prefix}_agent"
ops_schema: "string (required)" # default: "${user_schema_prefix}_ops"
warehouse_id: "string (required)" # SQL warehouse for DDL execution
required_volumes: # optional — defaults below if not provided
- { name: "knowledge_sources", schema: "agent", comment: "KA + retrieval source files" }
- { name: "agent_outputs", schema: "agent", comment: "Tool-generated artifacts (CSV, charts)" }
- { name: "signoffs", schema: "ops", comment: "SDLC 04b stakeholder + engineering signoff decision.md artifacts" }
# AgentSpec.required_volumes[] is appended verbatim
agent_schema / ops_schema may be overridden to any name that starts with ${user_schema_prefix}_; a non-prefixed override raises ValueError, because RULE_10 sanctions in-session provisioning of the participant's own prefixed schemas only.
Operations
flowchart TD
start["F0 invoked"] --> grant["Verify USE CATALOG + CREATE SCHEMA"]
grant --> agentSch["CREATE SCHEMA IF NOT EXISTS<br/>${catalog}.${agent_schema}"]
grant --> opsSch["CREATE SCHEMA IF NOT EXISTS<br/>${catalog}.${ops_schema}"]
agentSch --> volLoop["For each required_volume:<br/>create MANAGED volume idempotently"]
opsSch --> volLoop
volLoop --> capture["Capture { agent_schema, ops_schema, uc_volumes{name → path} }"]
Reference implementation
import time
from databricks.sdk import WorkspaceClient
from databricks.sdk.service.sql import StatementState
def provision_uc_resources(
*,
uc_catalog: str,
user_schema_prefix: str,
agent_schema: str,
ops_schema: str,
warehouse_id: str,
required_volumes: list[dict] | None = None,
ddl_timeout_s: int = 300,
) -> dict:
w = WorkspaceClient()
# F0 provisions only the participant's own prefixed schemas (RULE_10 foundation carve-out).
suffixes = {}
for role, schema in (("agent", agent_schema), ("ops", ops_schema)):
if not schema.startswith(f"{user_schema_prefix}_") or schema == f"{user_schema_prefix}_":
raise ValueError(
f"{role}_schema {schema!r} must start with '{user_schema_prefix}_': RULE_10 "
"sanctions in-session provisioning of the participant's own prefixed schemas only"
)
suffixes[role] = schema.removeprefix(f"{user_schema_prefix}_")
def run_ddl(statement: str) -> None:
"""Run one DDL statement to a terminal state; raise unless it SUCCEEDED (fail closed)."""
resp = w.statement_execution.execute_statement(
warehouse_id=warehouse_id, statement=statement, wait_timeout="30s",
)
deadline = time.monotonic() + ddl_timeout_s
while resp.status.state in (StatementState.PENDING, StatementState.RUNNING):
if time.monotonic() > deadline:
w.statement_execution.cancel_execution(resp.statement_id)
raise TimeoutError(f"DDL still {resp.status.state.value} after {ddl_timeout_s}s: {statement}")
time.sleep(2)
resp = w.statement_execution.get_statement(resp.statement_id)
if resp.status.state != StatementState.SUCCEEDED: # FAILED / CANCELED / CLOSED
error = resp.status.error.message if resp.status.error else "no error message"
raise RuntimeError(f"DDL {resp.status.state.value}: {error} ({statement})")
# 1. Schemas (DDL via statement execution against the workshop warehouse)
for suffix in suffixes.values():
run_ddl(
f"CREATE SCHEMA IF NOT EXISTS {uc_catalog}.{user_schema_prefix}_{suffix} "
f"COMMENT 'F0-managed: agent or ops assets'"
)
# 2. MANAGED volumes (DDL IF NOT EXISTS through the same warehouse)
defaults = [
{"name": "knowledge_sources", "schema": "agent",
"comment": "KA + retrieval source files"},
{"name": "agent_outputs", "schema": "agent",
"comment": "Tool-generated artifacts (CSV, charts)"},
{"name": "signoffs", "schema": "ops",
"comment": "SDLC 04b stakeholder + engineering signoff decision.md artifacts"},
]
volumes_spec = (required_volumes or []) + defaults
seen, paths = set(), {}
for v in volumes_spec:
suffix = suffixes["agent" if v["schema"] == "agent" else "ops"]
target_schema, name = f"{user_schema_prefix}_{suffix}", v["name"]
key = (target_schema, name)
if key in seen:
continue
seen.add(key)
comment = v.get("comment", "F0-managed volume").replace("'", "'")
run_ddl(
f"CREATE VOLUME IF NOT EXISTS {uc_catalog}.{user_schema_prefix}_{suffix}.{name} "
f"COMMENT '{comment}'"
) # raises unless SUCCEEDED, so only confirmed volumes are captured
paths[name] = f"/Volumes/{uc_catalog}/{target_schema}/{name}"
return {
"agent_schema": agent_schema,
"ops_schema": ops_schema,
"uc_volumes": paths,
}
The skill caller passes the result back to vibecoding-state exit so
downstream skills can read state://Resources.uc_volumes.knowledge_sources etc.
Outputs (handoff)
| Key |
Consumer |
Value |
agent_schema |
F2, F5, SDLC 02/04, Track A 05 (memory) |
${user_schema_prefix}_agent |
ops_schema |
F2 (OTeL tables), SDLC 07 (monitoring), SDLC 04b (signoffs) |
${user_schema_prefix}_ops |
uc_volumes.knowledge_sources |
F5 (Step 5_0 — branch B/C upload), F3 (Vector Search MCP source dir) |
/Volumes/${catalog}/${agent_schema}/knowledge_sources |
uc_volumes.agent_outputs |
Track A 03 csv_export @function_tool, any code-interpreter output |
/Volumes/${catalog}/${agent_schema}/agent_outputs |
uc_volumes.signoffs (alias signoff_volume_path) |
SDLC 04b (writes decision.md), SDLC 06 (reads it as the promotion gate) |
/Volumes/${catalog}/${ops_schema}/signoffs |
uc_volumes.<custom> |
Whichever consuming skill declared the volume |
/Volumes/${catalog}/${schema}/${name} |
Validation Gate
Delegation Map
| Need |
Delegate To |
| Bronze schema + Delta tables |
bronze/00-bronze-layer-setup |
OTeL trace tables in ${ops_schema} |
F2 |
KA source markdown into knowledge_sources |
F5 Step 5_0 — branch (A) skip if non-empty / (B) upload ka_source / (C) auto-generate from PRD |
| Lakebase memory tables |
Track A 05 |
Benchmark tables in ${agent_schema} |
SDLC 02 |
Monitoring tables in ${ops_schema} |
SDLC 07 |
Limitations to Plan For
- Catalog must exist. F0 does not create catalogs (org-level admin action).
- Workspace warehouse required. Schema DDL goes through the SQL warehouse,
not the Python serverless API.
- No drop semantics. F0 is create-only — to remove a schema or volume,
delete out of band; do not put
DROP in any agent skill.
- Volume size budgeting. Managed volumes inherit the catalog's storage
location; if your
agent_outputs may grow large, document a retention
policy in the consuming skill.
Notes to Carry Forward
| Key |
Produced By |
Value |
agent_schema |
F0 |
${user_schema_prefix}_agent |
ops_schema |
F0 |
${user_schema_prefix}_ops |
uc_volumes |
F0 |
Map of name → /Volumes/${catalog}/${schema}/${name} |
knowledge_source_path |
F0 (alias of uc_volumes.knowledge_sources) |
Path that F5 attaches as a files knowledge source |
agent_outputs_path |
F0 (alias of uc_volumes.agent_outputs) |
Path that Track A 03's csv_export writes to |
signoff_volume_path |
F0 (alias of uc_volumes.signoffs) |
Generic, use-case-neutral signoff volume path SDLC 04b writes decision.md into and SDLC 06 reads as the promotion gate. Always /Volumes/${catalog}/${ops_schema}/signoffs — never workshop-specific. |
Next Step
After F0's gate passes, continue with F1: MLflow GenAI Foundation. Every subsequent skill can now assume schemas + volumes exist.
References
Version History
| Version |
Date |
Changes |
| 1.1.0 |
2026-04-26 |
Added generic, workshop-neutral signoffs MANAGED volume in ${ops_schema} as a default required_volume. SDLC 04b writes decision.md artifacts here and SDLC 06 reads them as the promotion gate. Captures signoff_volume_path (alias of uc_volumes.signoffs) so downstream skills resolve /Volumes/${catalog}/${ops_schema}/signoffs/v<N>/decision.md deterministically without per-use-case naming. |
| 1.0.0 |
2026-04-24 |
Initial skill — extracted from F5 (05-knowledge-assistant) Step 5_0 so schema + volume provisioning is owned in one cross-cutting place. Consumers (F5, F2, Track A 05, SDLC 02/04/07) now delegate. |
1---2name: 00-uc-resources-foundation3description: >4 Use when any agent or downstream skill needs Unity Catalog schemas and managed5 volumes (knowledge_sources, agent_outputs, memory tables, benchmark tables,6 etc.) provisioned idempotently. Foundation Step 0 — runs before MLflow7 tracing setup and before any track-specific skill. Owned by no track. Reads8 the resolved spec to discover which volumes a use case needs and creates them9 alongside the canonical agent + ops UC schemas.10license: Apache-2.011clients: [ide_cli, genie_code]12bundle_resource: schemas13deploy_verb: bundle_deploy14deploy_note: "Provisions UC schemas + managed volumes idempotently under the per-user prefixed schema (`{user_schema_prefix}`); created via DDL `IF NOT EXISTS` identically on both clients. On Genie Code run CLI steps through runDatabricksCli (pre-authenticated) and execute on serverless; see `skills/genie-code-environment`."15coverage: full16metadata:17 source: https://skillmd.com/skills/databricks-solutions/00-uc-resources-foundation18 last_verified: "2026-04-15"19 volatility: medium20 upstream_sources: []21 author: "prashanth-subrahmanyam"22 version: "1.1.0"23 domain: "genai-agents"24 pipeline_position: "F0"25 consumes: "uc_catalog, USER_SCHEMA_PREFIX, AgentSpec.required_volumes(optional), AgentSpec.knowledge_base_backend(optional)"26 produces: "agent_schema, ops_schema, uc_volumes(map of name → /Volumes path), signoff_volume_path"27 grounded_in: "docs.databricks.com/aws/en/sql/language-manual/sql-ref-syntax-ddl-create-schema, docs.databricks.com/aws/en/sql/language-manual/sql-ref-syntax-ddl-create-volume, docs.databricks.com/aws/en/volumes/managed-vs-external"28fields_read:29 - resources.uc.catalog30 - resources.uc.user_schema_prefix31 - agent.required_volumes32 - agent.knowledge_base_backend33---3435# Foundation Step 0: UC Resources (Schemas + Volumes)3637> **Cross-cutting.** This skill is consumed by **every** Foundation step and38> agent track. It exists so that no downstream skill (KA, agent memory, tool39> exports, benchmark persistence, prompt registry) has to repeat schema or40> volume provisioning. If you find yourself writing `CREATE SCHEMA` or41> `CREATE VOLUME` inside another skill, delegate to F0 instead.4243## What This Skill Owns4445| Resource | Purpose | Default name |46|---|---|---|47| **Agent UC schema** | Holds agent-owned assets — KA source volume, tool output volume, benchmark tables, prompt registry pointers, memory tables | `${user_schema_prefix}_agent` |48| **Ops UC schema** | Holds operational assets — MLflow OTeL trace tables (F2), monitoring tables (SDLC 07), eval-run rollups (SDLC 04) | `${user_schema_prefix}_ops` |49| **`knowledge_sources` MANAGED volume** (in agent schema) | Source markdown / PDFs for Knowledge Assistants (F5) and any retrieval-style tool | `/Volumes/${catalog}/${agent_schema}/knowledge_sources` |50| **`agent_outputs` MANAGED volume** (in agent schema) | CSV exports, charts, generated artifacts produced by `@function_tool` calls | `/Volumes/${catalog}/${agent_schema}/agent_outputs` |51| **`signoffs` MANAGED volume** (in ops schema) | Stakeholder + engineering sign-off `decision.md` artifacts (SDLC 04b) — the durable, UC-governed location read by SDLC 06 promotion gates | `/Volumes/${catalog}/${ops_schema}/signoffs` |52| **Custom volumes** (in agent or ops schema) | Anything declared in `state://AgentSpec.required_volumes[]` | `/Volumes/${catalog}/${schema}/${name}` |5354> **`signoffs` is generic.** F0 pre-creates the volume under a workshop-neutral55> name (`signoffs`) so SDLC 04b never has to provision its own UC location and56> SDLC 06 can resolve `/Volumes/${catalog}/${ops_schema}/signoffs/v<N>/decision.md`57> deterministically. The volume name does **not** carry any use-case-specific58> prefix (no `skyloyalty_signoffs`, no `${use_case_slug}_signoffs`); the59> per-version directory `v<N>/` carries the disambiguation.6061Every operation is **idempotent** (`CREATE SCHEMA IF NOT EXISTS`,62`CREATE VOLUME IF NOT EXISTS`, both as DDL — an existing object is a no-op). It63is safe to call this skill from any prompt that needs to be sure these64resources exist; the cost on a warm workspace is one round-trip per resource.6566## What This Skill Does NOT Own6768- **Catalog creation.** A UC catalog must already exist; F0 reads69 `state://Resources.uc.catalog` and validates the caller has `USE CATALOG` +70 `CREATE SCHEMA` on it.71- **UC table DDL.** Tables for OTeL traces, benchmarks, memory, or monitoring72 are owned by their respective consuming skills (F2, SDLC 02/04/07).73- **Source content.** F0 creates an empty `knowledge_sources` volume; F5 (or74 any other consumer) is responsible for populating it with files.75- **Bronze data.** Bronze schema + Delta tables are owned by the bronze layer76 setup skill ([`data_product_accelerator/skills/bronze/00-bronze-layer-setup`](../../../data_product_accelerator/skills/bronze/00-bronze-layer-setup/SKILL.md)).7778## When to Use7980- **Always**, near the top of the prompt sequence — before F1 (MLflow env) so81 that anything downstream can assume schemas + volumes exist.82- Whenever you add a new consuming skill that needs a new volume — extend83 `state://AgentSpec.required_volumes[]` rather than putting `CREATE VOLUME`84 inside the consuming skill.85- After `bootstrap` when the resolved spec is available, but before any other86 Foundation step.8788## Prerequisites8990| Requirement | How to Check |91|---|---|92| `state://Resources.uc.catalog` is resolved | Set by `vibecoding-state` `bootstrap` op |93| `state://Resources.uc.user_schema_prefix` is resolved | Derived in `bootstrap` from the user identity + `use_case_slug` |94| Workshop SQL warehouse is RUNNING | `state://Resources.warehouse_id` (set by Prompt 1 preflight) — needed for `CREATE SCHEMA` execution via `WorkspaceClient.statement_execution` |95| `databricks-sdk >= 0.30` installed | `python -c "from databricks.sdk import WorkspaceClient; from databricks.sdk.service.sql import StatementState"` |96| Caller has `USE CATALOG` + `CREATE SCHEMA` on the catalog | `databricks unity-catalog catalogs get $CATALOG --output json \| jq '.privileges'` |9798---99100## Inputs101102```yaml103# Resolved by the caller from state — F0 does not read state directly.104uc_catalog: "string (required)" # e.g. "main"105user_schema_prefix: "string (required)" # e.g. "jane_d_booking_app"106agent_schema: "string (required)" # default: "${user_schema_prefix}_agent"107ops_schema: "string (required)" # default: "${user_schema_prefix}_ops"108warehouse_id: "string (required)" # SQL warehouse for DDL execution109required_volumes: # optional — defaults below if not provided110 - { name: "knowledge_sources", schema: "agent", comment: "KA + retrieval source files" }111 - { name: "agent_outputs", schema: "agent", comment: "Tool-generated artifacts (CSV, charts)" }112 - { name: "signoffs", schema: "ops", comment: "SDLC 04b stakeholder + engineering signoff decision.md artifacts" }113 # AgentSpec.required_volumes[] is appended verbatim114```115116`agent_schema` / `ops_schema` may be overridden to any name that starts with `${user_schema_prefix}_`; a non-prefixed override raises `ValueError`, because RULE_10 sanctions in-session provisioning of the participant's own prefixed schemas only.117118## Operations119120```mermaid121flowchart TD122 start["F0 invoked"] --> grant["Verify USE CATALOG + CREATE SCHEMA"]123 grant --> agentSch["CREATE SCHEMA IF NOT EXISTS<br/>${catalog}.${agent_schema}"]124 grant --> opsSch["CREATE SCHEMA IF NOT EXISTS<br/>${catalog}.${ops_schema}"]125 agentSch --> volLoop["For each required_volume:<br/>create MANAGED volume idempotently"]126 opsSch --> volLoop127 volLoop --> capture["Capture { agent_schema, ops_schema, uc_volumes{name → path} }"]128```129130### Reference implementation131132```python133import time134135from databricks.sdk import WorkspaceClient136from databricks.sdk.service.sql import StatementState137138def provision_uc_resources(139 *,140 uc_catalog: str,141 user_schema_prefix: str,142 agent_schema: str,143 ops_schema: str,144 warehouse_id: str,145 required_volumes: list[dict] | None = None,146 ddl_timeout_s: int = 300,147) -> dict:148 w = WorkspaceClient()149150 # F0 provisions only the participant's own prefixed schemas (RULE_10 foundation carve-out).151 suffixes = {}152 for role, schema in (("agent", agent_schema), ("ops", ops_schema)):153 if not schema.startswith(f"{user_schema_prefix}_") or schema == f"{user_schema_prefix}_":154 raise ValueError(155 f"{role}_schema {schema!r} must start with '{user_schema_prefix}_': RULE_10 "156 "sanctions in-session provisioning of the participant's own prefixed schemas only"157 )158 suffixes[role] = schema.removeprefix(f"{user_schema_prefix}_")159160 def run_ddl(statement: str) -> None:161 """Run one DDL statement to a terminal state; raise unless it SUCCEEDED (fail closed)."""162 resp = w.statement_execution.execute_statement(163 warehouse_id=warehouse_id, statement=statement, wait_timeout="30s",164 )165 deadline = time.monotonic() + ddl_timeout_s166 while resp.status.state in (StatementState.PENDING, StatementState.RUNNING):167 if time.monotonic() > deadline:168 w.statement_execution.cancel_execution(resp.statement_id)169 raise TimeoutError(f"DDL still {resp.status.state.value} after {ddl_timeout_s}s: {statement}")170 time.sleep(2)171 resp = w.statement_execution.get_statement(resp.statement_id)172 if resp.status.state != StatementState.SUCCEEDED: # FAILED / CANCELED / CLOSED173 error = resp.status.error.message if resp.status.error else "no error message"174 raise RuntimeError(f"DDL {resp.status.state.value}: {error} ({statement})")175176 # 1. Schemas (DDL via statement execution against the workshop warehouse)177 for suffix in suffixes.values():178 run_ddl(179 f"CREATE SCHEMA IF NOT EXISTS {uc_catalog}.{user_schema_prefix}_{suffix} "180 f"COMMENT 'F0-managed: agent or ops assets'"181 )182183 # 2. MANAGED volumes (DDL IF NOT EXISTS through the same warehouse)184 defaults = [185 {"name": "knowledge_sources", "schema": "agent",186 "comment": "KA + retrieval source files"},187 {"name": "agent_outputs", "schema": "agent",188 "comment": "Tool-generated artifacts (CSV, charts)"},189 {"name": "signoffs", "schema": "ops",190 "comment": "SDLC 04b stakeholder + engineering signoff decision.md artifacts"},191 ]192 volumes_spec = (required_volumes or []) + defaults193 seen, paths = set(), {}194 for v in volumes_spec:195 suffix = suffixes["agent" if v["schema"] == "agent" else "ops"]196 target_schema, name = f"{user_schema_prefix}_{suffix}", v["name"]197 key = (target_schema, name)198 if key in seen:199 continue200 seen.add(key)201 comment = v.get("comment", "F0-managed volume").replace("'", "'")202 run_ddl(203 f"CREATE VOLUME IF NOT EXISTS {uc_catalog}.{user_schema_prefix}_{suffix}.{name} "204 f"COMMENT '{comment}'"205 ) # raises unless SUCCEEDED, so only confirmed volumes are captured206 paths[name] = f"/Volumes/{uc_catalog}/{target_schema}/{name}"207208 return {209 "agent_schema": agent_schema,210 "ops_schema": ops_schema,211 "uc_volumes": paths,212 }213```214215The skill caller passes the result back to `vibecoding-state` `exit` so216downstream skills can read `state://Resources.uc_volumes.knowledge_sources` etc.217218---219220## Outputs (handoff)221222| Key | Consumer | Value |223|---|---|---|224| `agent_schema` | F2, F5, SDLC 02/04, Track A 05 (memory) | `${user_schema_prefix}_agent` |225| `ops_schema` | F2 (OTeL tables), SDLC 07 (monitoring), SDLC 04b (signoffs) | `${user_schema_prefix}_ops` |226| `uc_volumes.knowledge_sources` | F5 (Step 5_0 — branch B/C upload), F3 (Vector Search MCP source dir) | `/Volumes/${catalog}/${agent_schema}/knowledge_sources` |227| `uc_volumes.agent_outputs` | Track A 03 `csv_export` `@function_tool`, any code-interpreter output | `/Volumes/${catalog}/${agent_schema}/agent_outputs` |228| `uc_volumes.signoffs` (alias `signoff_volume_path`) | SDLC 04b (writes `decision.md`), SDLC 06 (reads it as the promotion gate) | `/Volumes/${catalog}/${ops_schema}/signoffs` |229| `uc_volumes.<custom>` | Whichever consuming skill declared the volume | `/Volumes/${catalog}/${schema}/${name}` |230231---232233## Validation Gate234235- [ ] `${uc_catalog}.${agent_schema}` exists (`SHOW SCHEMAS IN ${uc_catalog} LIKE '${agent_schema}'`)236- [ ] `${uc_catalog}.${ops_schema}` exists237- [ ] Every volume in `required_volumes` is listable via `databricks volumes list ${uc_catalog} ${schema} --output json | jq '.volumes[] | select(.name=="<name>")'`238- [ ] The generic `signoffs` volume in `${ops_schema}` is present (consumed by SDLC 04b / SDLC 06 — name is workshop-neutral)239- [ ] `agent_schema`, `ops_schema`, the full `uc_volumes` map, and `signoff_volume_path` are captured in state240241---242243## Delegation Map244245| Need | Delegate To |246|---|---|247| Bronze schema + Delta tables | [`bronze/00-bronze-layer-setup`](../../../data_product_accelerator/skills/bronze/00-bronze-layer-setup/SKILL.md) |248| OTeL trace tables in `${ops_schema}` | [F2](../02-experiment-tracing-and-uc-storage/SKILL.md) |249| KA source markdown into `knowledge_sources` | [F5 Step 5_0](../05-knowledge-assistant/SKILL.md) — branch (A) skip if non-empty / (B) upload `ka_source` / (C) auto-generate from PRD |250| Lakebase memory tables | [Track A 05](../../tracks/A-custom-agent-apps/05-memory-and-state/SKILL.md) |251| Benchmark tables in `${agent_schema}` | [SDLC 02](../../sdlc/02-evaluation-datasets/SKILL.md) |252| Monitoring tables in `${ops_schema}` | [SDLC 07](../../sdlc/07-production-monitoring/SKILL.md) |253254---255256## Limitations to Plan For257258- **Catalog must exist.** F0 does not create catalogs (org-level admin action).259- **Workspace warehouse required.** Schema DDL goes through the SQL warehouse,260 not the Python serverless API.261- **No drop semantics.** F0 is create-only — to remove a schema or volume,262 delete out of band; do not put `DROP` in any agent skill.263- **Volume size budgeting.** Managed volumes inherit the catalog's storage264 location; if your `agent_outputs` may grow large, document a retention265 policy in the consuming skill.266267---268269## Notes to Carry Forward270271| Key | Produced By | Value |272|---|---|---|273| `agent_schema` | F0 | `${user_schema_prefix}_agent` |274| `ops_schema` | F0 | `${user_schema_prefix}_ops` |275| `uc_volumes` | F0 | Map of `name → /Volumes/${catalog}/${schema}/${name}` |276| `knowledge_source_path` | F0 (alias of `uc_volumes.knowledge_sources`) | Path that F5 attaches as a `files` knowledge source |277| `agent_outputs_path` | F0 (alias of `uc_volumes.agent_outputs`) | Path that Track A 03's `csv_export` writes to |278| `signoff_volume_path` | F0 (alias of `uc_volumes.signoffs`) | Generic, use-case-neutral signoff volume path SDLC 04b writes `decision.md` into and SDLC 06 reads as the promotion gate. Always `/Volumes/${catalog}/${ops_schema}/signoffs` — never workshop-specific. |279280---281282## Next Step283284After F0's gate passes, continue with [F1: MLflow GenAI Foundation](../01-mlflow-genai-foundation/SKILL.md). Every subsequent skill can now assume schemas + volumes exist.285286## References287288- [CREATE SCHEMA — Databricks SQL](https://docs.databricks.com/aws/en/sql/language-manual/sql-ref-syntax-ddl-create-schema)289- [CREATE VOLUME — Databricks SQL](https://docs.databricks.com/aws/en/sql/language-manual/sql-ref-syntax-ddl-create-volume)290- [Managed vs External Volumes](https://docs.databricks.com/aws/en/volumes/managed-vs-external)291- [Databricks SDK for Python — `volumes`](https://databricks-sdk-py.readthedocs.io/en/latest/workspace/catalog/volumes.html)292293---294295## Version History296297| Version | Date | Changes |298|---------|------|---------|299| 1.1.0 | 2026-04-26 | Added generic, workshop-neutral `signoffs` MANAGED volume in `${ops_schema}` as a default `required_volume`. SDLC 04b writes `decision.md` artifacts here and SDLC 06 reads them as the promotion gate. Captures `signoff_volume_path` (alias of `uc_volumes.signoffs`) so downstream skills resolve `/Volumes/${catalog}/${ops_schema}/signoffs/v<N>/decision.md` deterministically without per-use-case naming. |300| 1.0.0 | 2026-04-24 | Initial skill — extracted from F5 (`05-knowledge-assistant`) Step 5_0 so schema + volume provisioning is owned in one cross-cutting place. Consumers (F5, F2, Track A 05, SDLC 02/04/07) now delegate. |