Topic · NASA-STD-8739.8

What is NASA-STD-8739.8, and how do I keep the software assurance evidence true in CI?

Gist

NASA-STD-8739.8 is the agency standard for software assurance, software safety, and independent verification and validation. The useful artefact is a verification report whose rows still match this graph. Proof prints that report from the current traces with proof doc generate verification-report, then fails the merge when the graph is stale.

proof doc generate verification-report --output docs/verification.html

Keep OSMA IV&V if you are a NASA programme. Keep the Class A through E classification. Proof is not that office, and a named reviewer is not technical, managerial, and financial independence.

Request demo Last quarter's pack vs this push

01 · The dated assurance pack

An SRS is not assurance. Last quarter's IV&V memo is not this push.

NPR 7150.2D tells you how to engineer the software. 8739.8 tells you how to assure it. A person can fill a verification table once. The next commit does not reopen Word.

The NASA SRS artefact lives on NPR 7150.2D. That page renders npr7150-srs. This page is the overlay: per-requirement verification status, formalization, realizability, vacuity, review, traces, and obligations at the assurance level you actually set.

8739.8 wants every requirement to have a recorded verification method, and the result to be evidence. Proof's verification-report template is that table from loaded project data: formalization status, Kind2, consistency, vacuity, review status, trace coverage, obligations met versus required. It is not a Fairmont memo. It is not witnessed testing on the target.

proof validate --preflight
proof doc generate verification-report --output docs/verification.html
proof audit --fail-level warn

The first command rejects a sentence the templates cannot parse. The second renders HTML (or Markdown) from the current graph. PDF is a separate path and currently fails. The third is the standing gate. Generating the report before traces are refreshed is how last quarter's pack got stale.

02 · The exhibit

Same headings. A date, or this graph.

The Word file still has verification status. The graph moved. Click the tabs.

Word / SWEHB

SYS-REQ-042 verified, dated June
Method test, still in the appendix

Rows present

This push

The shall changed. Vacuity is now true. The pack did not rerun Kind2.

No report

Word / SWEHB

Still last quarter. Still a date.

Keep the IV&V path

Proof

Ask is every shall still evidenced
Gate audit --fail-level warn

Merge blocked if the report is thinner than the graph

Same headings. A date, or this graph. Click the tabs.

8739.8 asks	What Proof runs	What Proof loses
Software assurance	`proof audit --fail-level warn` on the loaded graph	Not OSMA. Not a Class A certificate. Jama still authors the shall.
Verification evidence	`proof doc generate verification-report` from current traces	Not witnessed testing. Tests are not executed. Keep `go test`.
IV&V independence	A named reviewer on the register, plus the standing gate	Not technical, managerial, and financial independence. Fairmont still owns that.
Software safety	Hazard links on the graph, when you wrote them	Not NASA-STD-8719.13. Not a hazard analysis workshop. That H1 lives on hazard analysis.

The report command loads the project, then writes per-requirement rows:

proof validate --preflight
proof doc generate verification-report --format html --output docs/verification.html
proof audit --fail-level warn

Unknown requirement language fails before the report. A thin graph makes a thin check. Treat the HTML as a view of the graph, not as a second source of assurance. We have not run this against a frozen 8739.8 corpus, and we have not scored it against an OSMA IV&V pack on the same branch. The loss is named, not scored.

03 · The honest loss

A green Proof audit is not IV&V.

NASA defines independence in three dimensions. Proof will not pretend a named reviewer is Fairmont.

Technical independence means the IV&V team did not write the code. Managerial independence means they are not in the same reporting chain. Financial independence means budget pressure cannot quietly shut the check down. A standing proof audit in your own CI is none of those three. Keep OSMA if the class requires it.

Classification still belongs to the programme. Class A through E, and the rule that the higher class wins, live in NPR 7150.2D. Proof will not upgrade a Class E component because it became operational. Software engineering technical authority and software assurance technical authority still have to agree.

The SRS still lives on NPR 7150.2D. The ISO document types live on ISO 29148. Jama still authors. The four jobs still live on DO-178C-style verification.

Request demo How the install is bought

04 · Nearby questions

What people type next.

What is NPR 7150.2D? NASA's software engineering procedure. The SRS artefact lives on that URL. This page is the assurance overlay, not a twin.

What is software assurance? The 8739.8 activity: planned, independent-enough checks that the software does what the shalls said. Proof prints the current rows. It does not replace the office.

What is NASA IV&V? The Independent Verification and Validation programme. Technical, managerial, and financial independence. Proof is not that programme.

What is NASA-STD-8719.13? Software safety, hazards, severity. Different H1. Hazard analysis.

What is CMMI? A process-maturity model. Proof does not assess your process. Keep the appraiser.

What is a requirements traceability matrix, and how do I keep it true? 8739.8 wants the links re-derived, not pasted. RTM.