You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Pipeline block ⑤ (Call in-graph: the compiled node + the "how subterranean" knob). Part of the compile-to-weights epic.
Context. stargraph has no node kind that runs a compiled small model under a minimal system prompt. The compiled model is an opaque router — in tension with the "no LLM routes" thesis and with #117 (learned-policy retrieval = core non-goal). Resolution: ship it opt-in + provenance-tagged, with a per-graph knob.
Proposal. New CompiledWorkflowNode(NodeBase) + a determinism knob.
execute(state, ctx): read dialogue/transient state, call the registry-pinned small model (block ④) with ONLY the minimal system prompt + turns, merge the assistant turn back into state. Stamp emitted facts ProvenanceBundle origin='llm' (fathom/_provenance.py:25) so opacity is explicit in the audit trail.
Wire the declared-but-unused Skill.system_prompt (skills/base.py:58) as the minimal prompt seam (today it is never injected at LLM-call time).
Knob (per-graph):fully-compiled (whole workflow → one compiled model, no runtime rules), hybrid (compile fluent legs, keep Fathom/CLIPS rules + provenance at audited decision hubs), rules-only (no compiled node). Default = hybrid/rules-only to honor ADR: name learned-policy (GRPO) retrieval an explicit core non-goal #117.
Where.
New src/stargraph/nodes/compiled_workflow.py subclassing NodeBase (nodes/base.py:45, execute:63). Zero core change via NodeSpec.kind='module.path:ClassName' (cli/run.py:_resolve_node_factory:260, _resolve_class_kind:240); NodeSpec.config (ir/_models.py:267) carries model_id/version/system_prompt.
Hybrid hub transitions stay as RuleSpec/goto resolved by fathom/_adapter.py. Skill.state_schema (skills/base.py:59) whitelists the channels the compiled node may write.
Acceptance criteria.
CompiledWorkflowNode runs a registry-pinned model under a minimal system prompt; emitted facts tagged origin='llm'.
In hybrid mode, transient state marshals cleanly across the compiled node via state_schema channels.
Caveat. Which legs are "safe to compile" vs "must stay audited" is a judgment call with no automatic test — the eval harness (linked) is the guard.
Source: arXiv:2605.22502v1 — "Compiling Agentic Workflows into LLM Weights" (subterranean agents). Part of the end-to-end compile-to-weights pipeline epic; file refs verified against current main. Determinism: per-graph knob (fully-compiled | hybrid | rules-only) defaults to honor #117 (no learned core router).
Pipeline block ⑤ (Call in-graph: the compiled node + the "how subterranean" knob). Part of the compile-to-weights epic.
Context. stargraph has no node kind that runs a compiled small model under a minimal system prompt. The compiled model is an opaque router — in tension with the "no LLM routes" thesis and with #117 (learned-policy retrieval = core non-goal). Resolution: ship it opt-in + provenance-tagged, with a per-graph knob.
Proposal. New
CompiledWorkflowNode(NodeBase)+ a determinism knob.execute(state, ctx): read dialogue/transient state, call the registry-pinned small model (block ④) with ONLY the minimal system prompt + turns, merge the assistant turn back into state. Stamp emitted factsProvenanceBundle origin='llm'(fathom/_provenance.py:25) so opacity is explicit in the audit trail.Skill.system_prompt(skills/base.py:58) as the minimal prompt seam (today it is never injected at LLM-call time).fully-compiled(whole workflow → one compiled model, no runtime rules),hybrid(compile fluent legs, keep Fathom/CLIPS rules + provenance at audited decision hubs),rules-only(no compiled node). Default = hybrid/rules-only to honor ADR: name learned-policy (GRPO) retrieval an explicit core non-goal #117.Where.
src/stargraph/nodes/compiled_workflow.pysubclassingNodeBase(nodes/base.py:45,execute:63). Zero core change viaNodeSpec.kind='module.path:ClassName'(cli/run.py:_resolve_node_factory:260,_resolve_class_kind:240);NodeSpec.config(ir/_models.py:267) carries model_id/version/system_prompt.RuleSpec/goto resolved byfathom/_adapter.py.Skill.state_schema(skills/base.py:59) whitelists the channels the compiled node may write.Acceptance criteria.
CompiledWorkflowNoderuns a registry-pinned model under a minimal system prompt; emitted facts taggedorigin='llm'.state_schemachannels.Caveat. Which legs are "safe to compile" vs "must stay audited" is a judgment call with no automatic test — the eval harness (linked) is the guard.
Source: arXiv:2605.22502v1 — "Compiling Agentic Workflows into LLM Weights" (subterranean agents). Part of the end-to-end compile-to-weights pipeline epic; file refs verified against current
main. Determinism: per-graph knob (fully-compiled | hybrid | rules-only) defaults to honor #117 (no learned core router).