docs(harness): pin per-stage profile override as workflow.call, not a mid-container switch
docs/prds/external-agent-harness.md's "Selection" says a stage may name a
different profile than its workflow's. Investigated before writing new
runtime code, per instruction not to build container-switching machinery:
a Kit task runs in one container provisioned for one profile (Container
layout), so a stage needing a different profile cannot be a mid-container
runner swap without inventing per-stage re-provisioning the PRD does not
ask for and Container layout argues against ("cost scales with the run,
not with how many agents it fans out to").
The requirement is already met by workflow.call plus per-task profile
resolution, both pre-existing: postgres.Store.StartCall stores the
callee's OWN Workflow name (not the caller's), and daemon.pinTaskProfile
resolves a task's profile purely from its own WorkflowDefinitionYAML.
Chaining these means a stage wanting a different profile is its own
callee workflow, started via workflow.call, exactly the way
event-automation.md already describes ("the callee runs as its own run
with its own agent profile").
No new production code, only a PRD clarification (dev-docs regenerated)
plus TestWorkflowCallGivesACalleeItsOwnProfile (daemon_test.go), which
closes the gap between two already-tested facts that nothing previously
stated together: TestWorkflowCallerConformance (staterpc) proves a
callee's Workflow field is independent of its caller's; this proves that
independence is what makes Kit-vs-image runner selection independent too,
composing a caller pinned to an image profile with a real StartCall-
created callee pinned to a Kit profile.
This commit is contained in:
1 parent
cc6d606de4
commit
2604cd4cf4
3 files changed
+64
-1
No files matched your search
@@ -328,7 +328,7 @@
|
||||
"url": "/dev/prds/external-agent-harness/",
|
||||
"title": "External agent harnesses",
|
||||
"section": "PRDs",
|
||||
"body": "# External agent harnesses\n\n**Status:** Approved\n**Authority:** `docs/architecture/agent-system.md` (agent execution as a data\nboundary); settles the harness non-goal deferred by\n`docs/prds/service-decomposition.md`.\n**Compounds with:** `docs/prds/execution-tree-state-machine.md` (a harness\nrun is one `agent` StepExecution), `docs/prds/binding-secret-encryption.md`\n(the at-rest envelope reused for harness credentials).\n\n## Decision\n\nA workflow stage can run on an external coding-agent CLI instead of the\nbuilt-in agent loop: Claude Code, Codex, Pi, OMP, GitHub Copilot CLI, or any\nother CLI packaged as a Docker Sandbox Kit v3. Each authenticates with an API\nkey or the operator's subscription login, and can carry the operator's own\nskills, plugins, subagents and instructions.\n\nTwo boundaries hold regardless of what the harness does:\n\n- **The real credential never enters the container.** The container sees\n sentinel values; archie's egress proxy substitutes the real credential on\n outbound requests and performs OAuth refresh on the host.\n- **Archie's workflow guarantees are enforced outside the harness.** Git,\n protected paths, read-only stages, gates and budgets are checked by archie\n around the process and against the worktree afterwards.\n\n## Kits\n\nA harness is packaged as a Kit v3 image: an OCI image whose\n`vnd.docker.sandbox.kit.descriptor` manifest annotation declares what the\nharness needs. Archie reads descriptors with the published\n`github.com/docker/sandbox-kit-spec/v3/spec` package and does not maintain\nits own descriptor format. Published Kits run unmodified.\n\nArchie is a partial Kit runtime. It implements these capability types:\n\n| Type | Archie behaviour |\n| ------------------- | ----------------------------------------------------------------------------------------- |\n| `credential@1` | Proxy-managed only, described below. |\n| `network-policy@1` | Enforced by the egress proxy. |\n| `volume@1` | A Docker volume keyed by WorkflowExecution, deleted with it. |\n| `lifecycle@1` | Install and startup hooks run on container create. |\n| `agent-sessions@1` | The headless `prompt` and `resume` invocations for the stage. |\n| `agent-skills@1` | The operator's skills store, mounted read-only whatever the Kit asks. |\n| `agent-context@1` | Archie's stage instructions, delivered where the Kit declares. |\n| `resources@1` | Container CPU and memory limits. |\n\nEvery other type fails closed: a required entry refuses the launch, an\noptional one is skipped and recorded on the step. `credential@1` with\n`passthrough: true` and any credential for a forge service are refused.\n\nThe operator's own setup (plugins, subagents, instructions, settings) is a\nmixin Kit the operator builds `FROM` or alongside the harness Kit. It is\nversioned image content, rewritten on every create, never state a run can\nmodify.\n\n## Contract\n\nA harness run is a second implementation of `agentexec.Runner`. It takes the\nsame `Request` and returns the same `Result`, so workflows, the daemon and\nthe State Store do not know which runner ran a stage.\n\n`agent-sessions@1` supplies the invocation. Archie supplies, per CLI, an\noutput adapter that translates the CLI's JSON output stream into\n`ToolCallReporter` calls, `Usage` and a session ID. A CLI with no adapter can\nrun stages with no capture tools, reporting no usage; its token budget falls\nback to wall clock.\n\nWorkflow definition validation rejects, at load time, a stage whose needs the\nselected harness cannot meet: capture tools without an adapter, or a gate\nretry without a `resume` verb.\n\n## Selection\n\nA harness is an agent profile. `docs/prds/event-automation.md` defines a\nprofile as the execution environment a workflow selects; a harness profile\nnames a Kit composition (the workload Kit pinned by digest, and zero or more\nmixin Kits) in place of a container image. Everything else about profiles\nholds: a workflow names its profile, a stage may name a different one, and a\nprofile holds no secrets.\n\n`archie-agent` builds the stage's runner from the stage's profile: the\nai-sdk agent loop for an image profile, the harness runner for a Kit\nprofile. Image profiles, and every workflow already written against them,\nare unchanged. A Kit profile named by no workflow is a validation error.\n\nProfiles are org resources in the State Store, so a Kit profile belongs to\none org, is granted to its workspaces like any profile, and applies without\na restart.\n\n## Organisations\n\n`docs/prds/Line truncated
|
||||
"body": "# External agent harnesses\n\n**Status:** Approved\n**Authority:** `docs/architecture/agent-system.md` (agent execution as a data\nboundary); settles the harness non-goal deferred by\n`docs/prds/service-decomposition.md`.\n**Compounds with:** `docs/prds/execution-tree-state-machine.md` (a harness\nrun is one `agent` StepExecution), `docs/prds/binding-secret-encryption.md`\n(the at-rest envelope reused for harness credentials).\n\n## Decision\n\nA workflow stage can run on an external coding-agent CLI instead of the\nbuilt-in agent loop: Claude Code, Codex, Pi, OMP, GitHub Copilot CLI, or any\nother CLI packaged as a Docker Sandbox Kit v3. Each authenticates with an API\nkey or the operator's subscription login, and can carry the operator's own\nskills, plugins, subagents and instructions.\n\nTwo boundaries hold regardless of what the harness does:\n\n- **The real credential never enters the container.** The container sees\n sentinel values; archie's egress proxy substitutes the real credential on\n outbound requests and performs OAuth refresh on the host.\n- **Archie's workflow guarantees are enforced outside the harness.** Git,\n protected paths, read-only stages, gates and budgets are checked by archie\n around the process and against the worktree afterwards.\n\n## Kits\n\nA harness is packaged as a Kit v3 image: an OCI image whose\n`vnd.docker.sandbox.kit.descriptor` manifest annotation declares what the\nharness needs. Archie reads descriptors with the published\n`github.com/docker/sandbox-kit-spec/v3/spec` package and does not maintain\nits own descriptor format. Published Kits run unmodified.\n\nArchie is a partial Kit runtime. It implements these capability types:\n\n| Type | Archie behaviour |\n| ------------------- | ----------------------------------------------------------------------------------------- |\n| `credential@1` | Proxy-managed only, described below. |\n| `network-policy@1` | Enforced by the egress proxy. |\n| `volume@1` | A Docker volume keyed by WorkflowExecution, deleted with it. |\n| `lifecycle@1` | Install and startup hooks run on container create. |\n| `agent-sessions@1` | The headless `prompt` and `resume` invocations for the stage. |\n| `agent-skills@1` | The operator's skills store, mounted read-only whatever the Kit asks. |\n| `agent-context@1` | Archie's stage instructions, delivered where the Kit declares. |\n| `resources@1` | Container CPU and memory limits. |\n\nEvery other type fails closed: a required entry refuses the launch, an\noptional one is skipped and recorded on the step. `credential@1` with\n`passthrough: true` and any credential for a forge service are refused.\n\nThe operator's own setup (plugins, subagents, instructions, settings) is a\nmixin Kit the operator builds `FROM` or alongside the harness Kit. It is\nversioned image content, rewritten on every create, never state a run can\nmodify.\n\n## Contract\n\nA harness run is a second implementation of `agentexec.Runner`. It takes the\nsame `Request` and returns the same `Result`, so workflows, the daemon and\nthe State Store do not know which runner ran a stage.\n\n`agent-sessions@1` supplies the invocation. Archie supplies, per CLI, an\noutput adapter that translates the CLI's JSON output stream into\n`ToolCallReporter` calls, `Usage` and a session ID. A CLI with no adapter can\nrun stages with no capture tools, reporting no usage; its token budget falls\nback to wall clock.\n\nWorkflow definition validation rejects, at load time, a stage whose needs the\nselected harness cannot meet: capture tools without an adapter, or a gate\nretry without a `resume` verb.\n\n## Selection\n\nA harness is an agent profile. `docs/prds/event-automation.md` defines a\nprofile as the execution environment a workflow selects; a harness profile\nnames a Kit composition (the workload Kit pinned by digest, and zero or more\nmixin Kits) in place of a container image. Everything else about profiles\nholds: a workflow names its profile, a stage may name a different one, and a\nprofile holds no secrets.\n\n`archie-agent` builds the stage's runner from the stage's profile: the\nai-sdk agent loop for an image profile, the harness runner for a Kit\nprofile. Image profiles, and every workflow already written against them,\nare unchanged. A Kit profile named by no workflow is a validation error.\n\nA stage naming a different profile than its workflow's is a separate run,\nnot a mid-container runner switch: a Kit task runs in one container\nprovisioned for one profile (Container layout), so a sLine truncated
|
||||
},
|
||||
{
|
||||
"url": "/dev/prds/identity-decision/",
|
||||
|
||||
@@ -85,6 +85,19 @@ ai-sdk agent loop for an image profile, the harness runner for a Kit
|
||||
profile. Image profiles, and every workflow already written against them,
|
||||
are unchanged. A Kit profile named by no workflow is a validation error.
|
||||
|
||||
A stage naming a different profile than its workflow's is a separate run,
|
||||
not a mid-container runner switch: a Kit task runs in one container
|
||||
provisioned for one profile (Container layout), so a stage that needs a
|
||||
different one is its own callee, started the way `workflow.call` already
|
||||
starts one (`docs/prds/event-automation.md` "Workflows call workflows"),
|
||||
with its own agent profile resolved off its own task row. This needs no
|
||||
mechanism beyond `workflow.call` plus per-task profile resolution
|
||||
(`daemon.pinTaskProfile`), both of which already treat every task's
|
||||
profile independently of any caller's -- see
|
||||
`TestWorkflowCallerConformance` (staterpc) for the callee's independent
|
||||
`Workflow` field and `TestWorkflowCallGivesACalleeItsOwnProfile` (daemon)
|
||||
for the Kit-vs-image runner selection that independence produces.
|
||||
|
||||
Profiles are org resources in the State Store, so a Kit profile belongs to
|
||||
one org, is granted to its workspaces like any profile, and applies without
|
||||
a restart.
|
||||
|
||||
@@ -2353,6 +2353,56 @@ func TestRunViaAgentCarriesTheWorkflowProfile(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
// TestWorkflowCallGivesACalleeItsOwnProfile is the harness PRD's "a stage may
|
||||
// name a different profile" (docs/prds/external-agent-harness.md, Selection):
|
||||
// there is no mid-container runner switch, and none is built here. A stage
|
||||
// wanting a different profile -- a Kit harness where the rest of the workflow
|
||||
// runs on the built-in loop -- is its own callee workflow, started the way
|
||||
// workflow.call already starts one (StartCall), with its own agent profile
|
||||
// resolved independently by pinTaskProfile off the callee's own task row.
|
||||
// TestWorkflowCallerConformance (staterpc) already proves the callee's
|
||||
// Workflow field is independent of the caller's; this proves that
|
||||
// independence is what makes Kit-vs-image runner selection independent too.
|
||||
func TestWorkflowCallGivesACalleeItsOwnProfile(t *testing.T) {
|
||||
d, s, _ := daemonWithNATS(t)
|
||||
cfg := d.Cfg.Get()
|
||||
cfg.Containers.Profiles = map[string]config.AgentProfile{
|
||||
"net": {Image: "agent-net:1"},
|
||||
"contain": {Kit: []string{"contain-kit@sha256:ab"}, Adapter: agentexec.AdapterClaudeCode},
|
||||
}
|
||||
d.Cfg.Set(cfg)
|
||||
d.WorkflowDefinitions = &workflowDefinitionsStub{collection: workflow.WorkflowDefinitionCollection{Definitions: []workflow.WorkflowDefinitionEntry{
|
||||
{ID: "investigate", YAML: "id: investigate\nprofile: net\nsteps:\n - type: implement.prepare\n"},
|
||||
{ID: "contain", YAML: "id: contain\nrepository: none\nprofile: contain\nsteps:\n - type: agent.run\n settings: {mission: contain it}\n"},
|
||||
}}, version: 1}
|
||||
ctx := context.Background()
|
||||
|
||||
if _, err := s.EnqueueIssue(ctx, "acme", "widget", 41, "t", "b", "", ""); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
caller, err := s.ClaimNext(ctx)
|
||||
if err != nil || caller == nil {
|
||||
t.Fatalf("claim caller: (%v, %v)", caller, err)
|
||||
}
|
||||
caller.Workflow = "investigate"
|
||||
|
||||
// Exactly the call a workflow.call step makes (workflow_call.go's
|
||||
// runWorkflowCall -> tc.Calls.StartCall), not a hand-rolled shortcut.
|
||||
callee, err := s.StartCall(ctx, caller.ID, "contain", nil)
|
||||
if err != nil || callee == nil {
|
||||
t.Fatalf("StartCall: (%+v, %v)", callee, err)
|
||||
}
|
||||
|
||||
callerProfile, ok := d.pinTaskProfile(ctx, caller)
|
||||
if !ok || callerProfile.IsKit() {
|
||||
t.Fatalf("caller profile = %+v, %v; want a non-Kit profile", callerProfile, ok)
|
||||
}
|
||||
calleeProfile, ok := d.pinTaskProfile(ctx, callee)
|
||||
if !ok || !calleeProfile.IsKit() || calleeProfile.Adapter != agentexec.AdapterClaudeCode {
|
||||
t.Fatalf("callee profile = %+v, %v; want the Kit profile its own workflow names", calleeProfile, ok)
|
||||
}
|
||||
}
|
||||
|
||||
// A container that dies mid-run (max-uptime reaper, OOM, crash) never answers
|
||||
// the taskrun request. The run must end with the task parked, not left
|
||||
// running with the worker blocked on a reply that cannot come.
|
||||
|
||||
Reference in new issue
Block a user