Add federal cost-allowability analysis component set and workflow

component_catalog_overrides.yaml

@@ -805,3 +805,331 @@ components:
           CostShare, AwardDeliverable, Modification, Terms, and
           ConflictOfInterest (see prompt.md). The six-block shape itself
           is repo-local.
+
+  expense-transaction-extraction:
+    output_contract:
+      format: json_object
+      scope: repo_local_normalization_contract
+      notes: >
+        Repo-local single-expense transaction record normalized from
+        receipt, invoice, purchase-order, p-card, or general-ledger
+        evidence. Input-normalization contract for the cost-allowability
+        analysis workflow; not a shared AI4RA-UDM schema.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 1 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_normalization_contract
+        notes: >
+          Fields align to sponsored-project transaction semantics
+          (vendor, amount, account coding, period) but the record is a
+          prompt-library normalization contract, not a shared UDM schema.
+    related_components:
+      - slug: regulated-activity-classifier
+        relationship: provides_input_to
+        notes: The normalized expense record is the classifier's primary input.
+
+  award-allowability-terms-extraction:
+    output_contract:
+      format: json_object
+      scope: repo_local_normalization_contract
+      notes: >
+        Repo-local record of the federal award terms that govern cost
+        allowability — period of performance, approved budget, caps and
+        exclusions, federal prior-approval triggers, indirect-cost
+        treatment, and institutional compliance-approval requirements.
+        Input-normalization contract for the cost-allowability analysis
+        workflow; not a shared AI4RA-UDM schema.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 1 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_normalization_contract
+        notes: >
+          Aligns to sponsored-project award and budget semantics. It
+          deliberately separates federal prior-approval triggers from
+          institutional IRB / IACUC / IBC compliance-approval
+          requirements; the record is a prompt-library normalization
+          contract, not a shared UDM schema.
+    related_components:
+      - slug: cost-award-terms-conformance-check
+        relationship: provides_input_to
+        notes: Supplies the caps, exclusions, and restrictions the conformance check evaluates.
+
+  regulated-activity-classifier:
+    output_contract:
+      format: json_object
+      scope: repo_local_detection_contract
+      notes: >
+        Repo-local classification record flagging whether an expense
+        implicates an activity requiring institutional compliance
+        oversight — human subjects (IRB), animals (IACUC), or biosafety
+        (IBC). Detection-layer contract for the cost-allowability
+        analysis workflow; not a shared AI4RA-UDM schema.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 1 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow; its
+        output routes the protocol-approval check in Step 2.
+      udm_alignment:
+        status: repo_local_detection_contract
+        notes: >
+          Aligns to research-compliance semantics; the per-regime
+          classification shape is a prompt-library contract and is not a
+          shared UDM schema.
+    related_components:
+      - slug: protocol-approval-allowability-check
+        relationship: gates
+        notes: The regime flags determine whether the protocol-approval check applies.
+
+  cost-period-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the cost
+        period-of-performance check (2 CFR 200.403(h), 200.309). Emits
+        the shared cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  cost-reasonableness-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the cost
+        reasonableness check (2 CFR 200.404). Emits the shared
+        cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  cost-allocability-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the cost
+        allocability check (2 CFR 200.405). Emits the shared
+        cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  cost-consistent-treatment-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the
+        consistent cost treatment check (2 CFR 200.403(d), 200.405(c)).
+        Emits the shared cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  cost-award-terms-conformance-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the award
+        terms conformance check (2 CFR 200.403(b)). Emits the shared
+        cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  cost-prior-approval-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the
+        federal prior approval check (2 CFR 200.407). Emits the shared
+        cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+    related_components:
+      - slug: protocol-approval-allowability-check
+        relationship: sibling_distinct_regime
+        notes: >
+          Covers federal sponsor prior approval (2 CFR 200.407);
+          protocol-approval-allowability-check covers institutional IRB /
+          IACUC / IBC approval, a distinct regulatory regime.
+
+  cost-documentation-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the cost
+        documentation check (2 CFR 200.403(g)). Emits the shared
+        cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  protocol-approval-allowability-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the
+        protocol approval allowability check, covering institutional IRB
+        (45 CFR 46), IACUC (PHS Policy / 9 CFR), and IBC (NIH Guidelines
+        for Recombinant or Synthetic Nucleic Acid Molecules / 42 CFR 73)
+        oversight. Emits the shared cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow; its
+        applicability is gated by the regulated-activity-classifier
+        output.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to research-compliance semantics; the finding object is
+          a prompt-library contract shared across the cost-allowability
+          check components, not a shared UDM schema.
+    related_components:
+      - slug: regulated-activity-classifier
+        relationship: gated_by
+        notes: The classifier's regime flags determine whether this check applies.
+      - slug: cost-prior-approval-check
+        relationship: sibling_distinct_regime
+        notes: >
+          Covers institutional IRB / IACUC / IBC approval;
+          cost-prior-approval-check covers federal sponsor prior
+          approval, a distinct regulatory regime.
+
+  cost-selected-item-check:
+    output_contract:
+      format: json_object
+      scope: repo_local_single_requirement_finding_contract
+      notes: >
+        Repo-local single-requirement allowability finding for the
+        selected items of cost check (2 CFR 200.421-200.476). Routes an
+        expense to its governing selected-item section and emits the
+        shared cost-allowability finding object.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Validate JSON output against schema.json. Canonical invocation
+        surface is prompt.md. Invoked as a Step 2 task of the
+        workflows/cost-allowability-analysis Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_single_requirement_finding_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          finding object is a prompt-library contract shared across the
+          cost-allowability check components, not a shared UDM schema.
+
+  cost-allowability-determination:
+    output_contract:
+      format: markdown_review
+      scope: repo_local_human_readable_determination_contract
+      notes: >
+        Repo-local human-readable Markdown allowability determination
+        for a single expense. Synthesizes the upstream single-requirement
+        check findings into one decision (Allowable, Potential issue,
+        Missing info, or Not allowable). No JSON output schema.
+    triad_integration:
+      evaluation_datasets: []
+      harness_notes: >
+        Canonical invocation surface is prompt.md. Score the Markdown
+        response against expected.md golden cases for the overall
+        decision, the conservative-decision rule (any not_allowable
+        finding forces an overall Not allowable), per-check coverage, the
+        compliance-oversight flag, and non-fabrication. Invoked as the
+        Step 3 output task of the workflows/cost-allowability-analysis
+        Vandalizer workflow.
+      udm_alignment:
+        status: repo_local_human_readable_determination_contract
+        notes: >
+          Aligns to sponsored-project cost-allowability semantics; the
+          human-readable determination shape is maintained only in
+          prompt-library and is not a shared UDM schema.
+    related_components:
+      - slug: regulated-activity-classifier
+        relationship: upstream_compliance_signal
+        notes: >
+          The determination surfaces a compliance-oversight flag when the
+          protocol-approval check, gated by the classifier, returns
+          anything other than not_applicable.

components/award-allowability-terms-extraction/CHANGELOG.md

@@ -0,0 +1,10 @@
+# Changelog
+
+All notable changes to this component. Versions follow semver: MAJOR for output-contract breaks, MINOR for backward-compatible additions, PATCH for wording or clarity.
+
+## [0.1.0] — 2026-05-21
+
+- Initial experimental release.
+- Structured award-terms record covering period of performance, approved budget categories, caps and exclusions, federal prior-approval triggers, indirect-cost treatment, compliance-approval requirements, and special conditions.
+- `prior_approval_triggers` (federal sponsor prior approval) and `compliance_approval_requirements` (institutional IRB / IACUC / IBC approvals) are kept as separate fields because they are distinct regulatory regimes.
+- No eval cases yet — status `experimental` until at least one golden extraction is added under `evals/cases/`.

components/award-allowability-terms-extraction/README.md

@@ -0,0 +1,41 @@
+# Award Allowability Terms Extraction
+
+Extracts the federal award terms that govern cost allowability — period of performance, approved budget categories, caps and exclusions, sponsor prior-approval triggers, indirect-cost treatment, and institutional compliance-approval requirements — into one structured record. It is an input-normalization step of the federal cost-allowability analysis workflow.
+
+**Current version:** 0.1.0
+**Category:** extraction
+**Domain:** research-administration
+**Status:** experimental
+**Manifestations:** prompt
+**Output contract:** [`schema.json`](schema.json)
+**Contract scope:** repo-local
+
+## Inputs
+
+A federal award notice, agreement, or terms-and-conditions document, as pasted text, an attachment, or a URL.
+
+## Outputs
+
+A single JSON object — see [`schema.json`](schema.json) — capturing the award-side constraints on allowability. Missing scalars are `null`; missing lists are empty arrays. The component records terms only; it does not review an expense.
+
+## Contract scope
+
+Repo-local. The record is a prompt-library normalization contract consumed by the downstream cost-allowability check components. It deliberately separates `prior_approval_triggers` (federal sponsor prior approval, 2 CFR 200.407-style) from `compliance_approval_requirements` (institutional IRB / IACUC / IBC approvals), because the two are distinct regulatory regimes. It is not a shared AI4RA-UDM schema.
+
+## Triad integration
+
+- **Evaluation datasets:** none yet — repo-local synthetic coverage planned.
+- **Harness notes:** canonical manifestation is `prompt.md`; validation surface is `schema.json`. Invoked as a Step 1 task of the [`cost-allowability-analysis`](https://github.com/AI4RA/prompt-library/tree/main/workflows/cost-allowability-analysis) workflow.
+- **Shared UDM relationship:** aligned to sponsored-project award and budget semantics; does not define or depend on a shared UDM schema.
+
+## Manifestations
+
+- [`prompt.md`](prompt.md) — canonical, LLM-agnostic prompt
+
+## Evals
+
+See [`evals/`](evals/).
+
+## Provenance
+
+Created 2026-05-21 as an input-normalization component of the federal cost-allowability analysis component set.

components/award-allowability-terms-extraction/evals/README.md

@@ -0,0 +1,13 @@
+# Evals — award-allowability-terms-extraction
+
+Each case lives under `cases/<case-slug>/` with at minimum `metadata.yaml`, `input-source.md`, `expected.json`, and optional `notes.md`. Run artifacts go under `runs/` (gitignored).
+
+## Planned cases
+
+- **NIH award with human-subjects terms** — exercises a populated `compliance_approval_requirements` array.
+- **NSF award with foreign-travel prior-approval term** — exercises `prior_approval_triggers`.
+- **Award with a dollar-capped budget category** — exercises `caps_and_exclusions` and a category `restrictions` value.
+
+## `validated_against_version`
+
+Every case must declare the component version that its `expected.json` was validated against.