Skip to content

BdM-15/capture-insights / skills/compliance-auditor

compliance-auditor

Verified

Federal acquisition compliance auditor for the active Theseus workspace, backed by live FAR/DFARS text via the vendored `ecfr` MCP. USE WHEN the user asks to audit FAR/DFARS clause coverage, validate that cited clauses actually exist in eCFR (catch fabricated or typo'd numbers), check whether a cited clause has been amended since the solicitation issued, validate regulatory references (NIST SP, DAFI, MIL-STD), check that every "shall" requirement has a deliverable, find missing compliance artifacts, audit proposal_instruction ↔ evaluation_factor coverage (UCF Section L↔M or non-UCF equivalent — FAR 16 task orders, FOPRs, BPA calls, OTAs), or "are we compliant with the proposal instructions?". Cross-references the workspace's clause / regulatory_reference / requirement / deliverable / compliance_artifact entities against live eCFR and flags gaps with severity. Format-agnostic. DO NOT USE FOR drafting compliant prose (use proposal-generator) or extracting clauses (Theseus pipeline does that automatically).

FileKindSize
SKILL.mdskill md13.0 KB
assets/finding.mdmarkdown1.8 KB
evals/evals.jsonconfig4.8 KB
references/cybersecurity_crosscut.mdmarkdown1.0 KB
references/ecfr_tools.mdmarkdown3.3 KB
references/far_dfars_quickref.mdmarkdown1.9 KB
references/severity_rubric.mdmarkdown1.1 KB

assets/finding.md

View on GitHub
# Finding Object Schema

Every finding emitted by the compliance-auditor MUST conform to this shape:

```json
{
  "id": "F-001",
  "check": "C3",
  "severity": "critical | high | medium | low | info",
  "entity_id": "<workspace entity id>",
  "entity_type": "<one of the 33 ontology entity types>",
  "evidence": "<verbatim text or paraphrase pointing at the gap>",
  "remediation": "<concrete next action — name the missing entity type or relationship>",
  "source_chunks": ["chunk-xxxxxxxx", "..."]
}
```

## Field semantics

- **`id`** — Sequential `F-001`, `F-002`, ... in emission order. Stable for a given run.
- **`check`** — Check ID (`C1`–`C8`) from SKILL.md.
- **`severity`** — From `references/severity_rubric.md`. Critical = will be rated Unacceptable; do not inflate.
- **`entity_id`** — The exact `entity_id` of the failing entity, as returned by `kg_entities` or `kg_query`. NEVER a free-form name.
- **`entity_type`** — Lowercase snake_case ontology type (e.g. `requirement`, `clause`, `evaluation_factor`).
- **`evidence`** — One sentence describing the gap. May quote chunk text in `"..."` or summarize. Must point at WHY the entity fails the check.
- **`remediation`** — One-line concrete action. Name the missing entity type or relationship (e.g. "Add a `SATISFIED_BY` edge to a `deliverable` entity").
- **`source_chunks`** — At least one `chunk_id` from `kg_chunks` or `kg_query` that justifies the finding. **Required for every finding.** No citation = no finding.

## Anti-patterns

- ❌ Aggregated findings ("12 requirements missing deliverables" as one row). Emit one finding per entity.
- ❌ Free-form `entity_id` like "the cybersecurity requirement". Use the actual entity ID.
- ❌ Empty `source_chunks`. If you cannot cite, you cannot raise.
- ❌ "Section L is missing" as evidence on a non-UCF solicitation. Map to entities, not headings.
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall