coverage: drop the rendered report section

The ledger stays a machine-written artifact (coverage.json and the SARIF
non-failing results); the executive report goes back to being the agent's
document.

strix/report/coverage.py

@@ -18,7 +18,7 @@ in, and they are kept apart on purpose:
 
 A coverage claim is an attestation, so conflating the two would be the worst
 possible failure: a hallucinated "tested and clean" is strictly less honest
-than no coverage section at all. Every entry therefore carries its ``source``,
+than no coverage record at all. Every entry therefore carries its ``source``,
 and machine-observed facts contradict rather than confirm — an agent that
 carried the ``sql_injection`` skill and recorded nothing about SQL injection
 shows up under ``gaps``, and a run that hit its budget ceiling is stamped
@@ -420,74 +420,3 @@ def write_coverage(run_dir: Path, document: dict[str, Any]) -> Path:
         len(document.get("gaps", [])),
     )
     return path
-
-
-def render_coverage_markdown(document: dict[str, Any]) -> str:
-    """Render the coverage appendix appended to the executive report.
-
-    Built from the document rather than written by an agent, so the table in
-    the client-facing report cannot drift from the ledger it summarizes.
-    """
-    summary = document.get("summary", {})
-    completeness = document.get("completeness", {})
-    entries = document.get("entries", [])
-    gaps = document.get("gaps", [])
-
-    lines = [
-        "# Coverage",
-        "",
-        "Surfaces assessed during this engagement and how each was closed. Entries are "
-        "reported by the testing agents; the completeness notes below are recorded by "
-        "the runtime.",
-        "",
-        f"**Surfaces reviewed:** {summary.get('surfaces_reviewed', 0)}  ",
-        f"**Findings filed:** {summary.get('findings_filed', 0)}  ",
-        f"**Open items and gaps:** {summary.get('gaps', 0)}",
-        "",
-    ]
-
-    if entries:
-        lines += [
-            "| Surface | Risk Area | Outcome | Evidence |",
-            "| --- | --- | --- | --- |",
-        ]
-        lines += [
-            "| {} | {} | {} | {} |".format(
-                _cell(entry.get("surface")),
-                _cell(entry.get("risk_area")),
-                _cell(entry.get("outcome_label") or entry.get("outcome")),
-                _cell(entry.get("evidence")),
-            )
-            for entry in entries
-        ]
-        lines.append("")
-    else:
-        lines += [
-            "No surfaces were recorded for this scan. The absence of findings below "
-            "cannot be read as evidence that any particular area was tested.",
-            "",
-        ]
-
-    if gaps:
-        lines += ["## Not Covered", ""]
-        lines += [f"- {_inline(gap.get('detail'))}" for gap in gaps]
-        lines.append("")
-
-    caveats = completeness.get("caveats") or []
-    if caveats:
-        lines += ["## Completeness", ""]
-        lines += [f"- {_inline(caveat)}" for caveat in caveats]
-        lines.append("")
-
-    return "\n".join(lines)
-
-
-def _cell(value: Any) -> str:
-    """Flatten *value* for a markdown table cell."""
-    text = _inline(value)
-    return text.replace("|", "\\|") if text else "—"
-
-
-def _inline(value: Any) -> str:
-    """Collapse *value* to a single line of plain text."""
-    return " ".join(str(value or "").split())

strix/report/state.py

@@ -14,7 +14,7 @@ from agents.usage import Usage
 from strix.config import codex
 from strix.config.loader import load_settings
 from strix.core.paths import run_dir_for, runtime_state_dir
-from strix.report.coverage import render_coverage_markdown, write_coverage
+from strix.report.coverage import write_coverage
 from strix.report.sarif import write_sarif
 from strix.report.usage import LLMUsageLedger
 from strix.report.writer import (
@@ -469,11 +469,7 @@ class ReportState:
                     logger.exception("coverage.json write failed (non-fatal)")
 
             if self.final_scan_result:
-                write_executive_report(
-                    run_dir,
-                    self.final_scan_result,
-                    render_coverage_markdown(coverage) if coverage else None,
-                )
+                write_executive_report(run_dir, self.final_scan_result)
 
             if self.vulnerability_reports:
                 write_vulnerabilities(run_dir, self.vulnerability_reports, self._saved_vuln_ids)

strix/report/writer.py

@@ -113,20 +113,12 @@ def write_run_record(run_dir: Path, run_record: dict[str, Any]) -> None:
     )
 
 
-def write_executive_report(
-    run_dir: Path,
-    final_scan_result: str,
-    coverage_markdown: str | None = None,
-) -> None:
+def write_executive_report(run_dir: Path, final_scan_result: str) -> None:
     path = run_dir / "penetration_test_report.md"
-    sections = [
-        "# Security Penetration Test Report\n",
-        f"**Generated:** {datetime.now(UTC).strftime('%Y-%m-%d %H:%M:%S UTC')}\n",
-        f"{final_scan_result}\n",
-    ]
-    if coverage_markdown:
-        sections.append(f"{coverage_markdown.rstrip()}\n")
-    atomic_write_text(path, "\n".join(sections))
+    with path.open("w", encoding="utf-8") as f:
+        f.write("# Security Penetration Test Report\n\n")
+        f.write(f"**Generated:** {datetime.now(UTC).strftime('%Y-%m-%d %H:%M:%S UTC')}\n\n")
+        f.write(f"{final_scan_result}\n")
     logger.info("Saved final penetration test report to: %s", path)
 
 

strix/tools/finish/tool.py

@@ -229,13 +229,7 @@ async def finish_scan(
         - ``methodology`` — frameworks followed (OWASP WSTG, PTES,
           OSSTMM, NIST), engagement type (black/gray/white box), scope
           and constraints, categories of testing performed. **No**
-          internal execution detail. Do **not** hand-write a table of
-          reviewed surfaces here: the report gains a ``Coverage``
-          section rendered directly from the coverage ledger, and a
-          transcribed copy would only drift from it. Describe the
-          approach and the classes of testing, and reconcile the ledger
-          via ``list_coverage`` (step 5) so that rendered section is
-          complete and honest.
+          internal execution detail.
         - ``technical_analysis`` — consolidated findings overview with
           severity model and systemic root causes. Reference individual
           vuln reports for repro steps; don't duplicate raw evidence.

tests/test_report_coverage.py

@@ -9,7 +9,6 @@ from strix.report.coverage import (
     _SKILL_PHRASINGS,
     build_coverage_document,
     read_agent_graph,
-    render_coverage_markdown,
     write_coverage,
 )
 from strix.skills import get_available_skills
@@ -171,27 +170,6 @@ def test_write_coverage_emits_a_top_level_artifact(tmp_path: Path) -> None:
     assert json.loads(path.read_text(encoding="utf-8"))["schema_version"] == 1
 
 
-def test_markdown_renders_a_surface_table() -> None:
-    markdown = render_coverage_markdown(_document())
-
-    assert "# Coverage" in markdown
-    assert "| POST /api/orders/{id} | object-level authorization | No issue identified |" in (
-        markdown
-    )
-
-
-def test_markdown_says_so_when_nothing_was_recorded() -> None:
-    markdown = render_coverage_markdown(_document(entries=[], agent_graph={}))
-
-    assert "cannot be read as evidence" in markdown
-
-
-def test_markdown_cell_escapes_pipes() -> None:
-    markdown = render_coverage_markdown(_document(entries=[_entry(surface="a|b")]))
-
-    assert "| a\\|b |" in markdown
-
-
 def test_read_agent_graph_tolerates_a_missing_or_corrupt_snapshot(tmp_path: Path) -> None:
     assert read_agent_graph(tmp_path) == {}
 

tests/test_report_writer.py

@@ -206,11 +206,3 @@ def test_render_vulnerability_md_surfaces_calibration_metadata() -> None:
     assert "## Confidence Rationale" in md
     assert "## What Would Change This Severity" in md
     assert "## Fix Verification" in md
-
-
-def test_write_executive_report_appends_the_coverage_section(tmp_path: Path) -> None:
-    write_executive_report(tmp_path, "Scan complete.", "# Coverage\n\nNothing tested.\n")
-    content = (tmp_path / "penetration_test_report.md").read_text(encoding="utf-8")
-
-    assert content.index("Scan complete.") < content.index("# Coverage")
-    assert "Nothing tested." in content

tests/test_state_coverage_artifact.py

@@ -45,17 +45,6 @@ def test_coverage_is_written_beside_the_other_artifacts(state: ReportState) -> N
     assert document["summary"]["surfaces_reviewed"] == 1
 
 
-def test_report_carries_a_coverage_section_from_the_ledger(state: ReportState) -> None:
-    _record_a_cleared_surface()
-    state.final_scan_result = "Scan complete."
-
-    state._save_artifacts()
-
-    report = (state.get_run_dir() / "penetration_test_report.md").read_text(encoding="utf-8")
-    assert "# Coverage" in report
-    assert "POST /api/orders/{id}" in report
-
-
 def test_cleared_surfaces_reach_sarif(state: ReportState) -> None:
     _record_a_cleared_surface()