fix: use platform llm grants for builder
All checks were successful
build / build (push) Successful in 2s
All checks were successful
build / build (push) Successful in 2s
This commit is contained in:
@@ -47,7 +47,7 @@ Three platform-managed bits land on the inner skill via `RunContext`:
|
|||||||
| Field | Source | Required? |
|
| Field | Source | Required? |
|
||||||
|---|---|---|
|
|---|---|---|
|
||||||
| `ctx.workspace.bucket` | grant minted by the orchestrator | yes |
|
| `ctx.workspace.bucket` | grant minted by the orchestrator | yes |
|
||||||
| `ctx.llm` | caller's LLM creds (Card declares `llm_provisioning=caller_provided`) | yes — the LLM bill goes to the caller |
|
| `ctx.llm` | platform-scoped LiteLLM grant (Card declares `llm_provisioning=platform`) | yes |
|
||||||
| `ctx.cp_jwt` | caller's CP JWT (Card declares `wants_cp_jwt=True`) | yes — used by `cp_deploy_tarball` |
|
| `ctx.cp_jwt` | caller's CP JWT (Card declares `wants_cp_jwt=True`) | yes — used by `cp_deploy_tarball` |
|
||||||
|
|
||||||
The user opts into all three when they pick `agent-builder` from the
|
The user opts into all three when they pick `agent-builder` from the
|
||||||
@@ -55,8 +55,8 @@ marketplace — the dashboard already surfaces these on the agent card.
|
|||||||
|
|
||||||
## Pricing
|
## Pricing
|
||||||
|
|
||||||
`$0.10 / call`, plus your LLM cost via your own provider. The deployed
|
`$0.10 / call` during the current pricing model. The deployed
|
||||||
agent is yours forever; subsequent invocations of YOUR agent run on
|
agent is yours forever; subsequent invocations of your agent run on
|
||||||
their own pricing (whatever you set when generating it).
|
their own pricing (whatever you set when generating it).
|
||||||
|
|
||||||
## How to call it
|
## How to call it
|
||||||
|
|||||||
13
agent.py
13
agent.py
@@ -10,8 +10,8 @@ This agent only works when invoked through the platform orchestrator,
|
|||||||
because it needs three things forwarded from the caller:
|
because it needs three things forwarded from the caller:
|
||||||
|
|
||||||
1. ``ctx.workspace.bucket`` — the user's MinIO bucket
|
1. ``ctx.workspace.bucket`` — the user's MinIO bucket
|
||||||
2. ``ctx.llm`` — caller's LLM creds (this is an expensive skill;
|
2. ``ctx.llm`` — a platform-scoped LiteLLM grant token for the builder's
|
||||||
the user pays for inference directly via their own provider)
|
own DeepAgents loop
|
||||||
3. ``ctx.cp_jwt`` — the user's CP JWT, so cp_deploy_tarball can
|
3. ``ctx.cp_jwt`` — the user's CP JWT, so cp_deploy_tarball can
|
||||||
POST to /v1/agents/from-tarball as the user.
|
POST to /v1/agents/from-tarball as the user.
|
||||||
"""
|
"""
|
||||||
@@ -53,15 +53,18 @@ class AgentBuilder(A2AAgent[BuilderConfig, NoAuth]):
|
|||||||
config_model = BuilderConfig
|
config_model = BuilderConfig
|
||||||
auth_model = NoAuth
|
auth_model = NoAuth
|
||||||
tools_used = ("deepagents", "langgraph", "a2a-pack", "litellm", "microsandbox")
|
tools_used = ("deepagents", "langgraph", "a2a-pack", "litellm", "microsandbox")
|
||||||
llm_provisioning = LLMProvisioning.CALLER_PROVIDED
|
llm_provisioning = LLMProvisioning.PLATFORM
|
||||||
wants_cp_jwt = True
|
wants_cp_jwt = True
|
||||||
workspace_access = WorkspaceAccess.dynamic(
|
workspace_access = WorkspaceAccess.dynamic(
|
||||||
allowed_modes=(WorkspaceMode.READ_ONLY, WorkspaceMode.READ_WRITE_OVERLAY),
|
allowed_modes=(WorkspaceMode.READ_ONLY, WorkspaceMode.READ_WRITE_OVERLAY),
|
||||||
)
|
)
|
||||||
pricing = Pricing(
|
pricing = Pricing(
|
||||||
price_per_call_usd=0.10,
|
price_per_call_usd=0.10,
|
||||||
caller_pays_llm=True,
|
caller_pays_llm=False,
|
||||||
notes="LLM cost passes through; deployed agent is yours forever.",
|
notes=(
|
||||||
|
"Uses a platform-scoped LiteLLM grant for the builder's own "
|
||||||
|
"reasoning loop; deployed agent is yours forever."
|
||||||
|
),
|
||||||
)
|
)
|
||||||
|
|
||||||
@skill(
|
@skill(
|
||||||
|
|||||||
@@ -27,8 +27,9 @@ class BuilderContext:
|
|||||||
project_prefix: str | None = None
|
project_prefix: str | None = None
|
||||||
workspace: Any | None = None
|
workspace: Any | None = None
|
||||||
grant_token: str | None = None
|
grant_token: str | None = None
|
||||||
# Optional caller-provided LLM creds (when the outer platform Card
|
# LLM creds forwarded by the orchestrator according to this agent's Card.
|
||||||
# declares llm_provisioning=caller_provided). Falls back to settings.
|
# The builder itself uses platform grants; generated agents usually use
|
||||||
|
# caller-provided creds. Falls back to settings for local development.
|
||||||
llm_base_url: str | None = None
|
llm_base_url: str | None = None
|
||||||
llm_api_key: str | None = None
|
llm_api_key: str | None = None
|
||||||
llm_model: str | None = None
|
llm_model: str | None = None
|
||||||
|
|||||||
@@ -24,6 +24,13 @@ class BuilderPromptTests(unittest.TestCase):
|
|||||||
)
|
)
|
||||||
self.assertIn("stream=True", agent_source)
|
self.assertIn("stream=True", agent_source)
|
||||||
|
|
||||||
|
def test_outer_builder_uses_platform_llm_grants(self) -> None:
|
||||||
|
agent_source = Path(__file__).resolve().parents[1].joinpath("agent.py").read_text()
|
||||||
|
|
||||||
|
self.assertIn("llm_provisioning = LLMProvisioning.PLATFORM", agent_source)
|
||||||
|
self.assertIn("caller_pays_llm=False", agent_source)
|
||||||
|
self.assertIn("platform-scoped LiteLLM grant", agent_source)
|
||||||
|
|
||||||
def test_builder_defaults_raise_timeout_budget(self) -> None:
|
def test_builder_defaults_raise_timeout_budget(self) -> None:
|
||||||
config_source = Path(__file__).resolve().parents[1].joinpath(
|
config_source = Path(__file__).resolve().parents[1].joinpath(
|
||||||
"agent_builder/config.py"
|
"agent_builder/config.py"
|
||||||
|
|||||||
Reference in New Issue
Block a user