ECSS Verification — Test Report DRD (space-systems/ecss/e1002-test-report)
Use when the task is to produce or check the test report of
ECSS-E-ST-10-02 clause 5.3.2.1 and Annex C -- the record that turns a
test run into verification evidence against a specific article, a
specific facility and a specific procedure revision.
Domain quick reference
- The report cites the test procedure by revision, and the cited
revision must be the one actually executed. This is not a
typographic class of defect: with the wrong revision the report
describes a test that was not run, and every conclusion in it is
attributed to the wrong method.
- An absent revision and a mismatched revision are handled by different
checks, so a blank field is reported once as missing identification
rather than twice.
- Identification is what lets data be attributed at all: article,
configuration, facility, procedure and revision. Data that cannot be
attributed to a specific article in a specific state is not evidence
about anything.
- Every test objective carries its own verdict. A report is not a pass
because most objectives passed -- each objective was written because
it had to be demonstrated.
- An objective marked not performed needs a reason. Without one, the
reader cannot distinguish a deliberate descope from an omission, and
those close very differently.
- Overall verdict is severity ordered: any failure fails the report;
otherwise anything not performed leaves it incomplete; only an
all-pass set passes.
- A discrepancy with no nonconformance raised against it is an
observation nobody is obliged to close. Linking it is what puts it
into a process that ends in a disposition.
- Summarized results need a reference to the recorded data. Re-analysis
is what makes test data evidence rather than testimony.
Workflow
- Check the identification fields; report each that is absent.
- Compare the cited procedure revision against the revision actually
run and report a mismatch.
- For each objective, confirm a verdict is recorded, and that a
not-performed objective carries its reason; reject duplicate
objective identifiers.
- For each discrepancy, confirm a nonconformance reference is present.
- Confirm the measured data reference is present.
- Aggregate the objective verdicts under the severity ordering.
- The report is acceptable only when it passes and no finding stands.
Pitfalls
- Citing the procedure without its revision, or carrying the revision
from the plan rather than from what the operators actually ran.
- Reporting an overall pass because the failing objective was "minor".
Severity belongs in the disposition of the nonconformance, not in the
arithmetic of the verdict.
- Marking an objective not performed and leaving the reason to the
reader, who cannot tell a descope from a miss.
- Recording a discrepancy in narrative text with no nonconformance
raised. It reads as disclosed and enters no process.
- Presenting summarized results with no pointer to the recorded data,
which makes independent re-analysis impossible.
- Letting two objectives share an identifier, so one verdict silently
overwrites another in any downstream roll-up.
Behavior contract (gate 3)
The identification, procedure-revision, objective-verdict,
discrepancy-linkage, data-reference and severity-ordered aggregation
logic is exercised by the gate 3 contract test:
scripts/test_e1002_test_report.py against
scripts/e1002_test_report_logic.py (stdlib unittest, offline). Run:
python3 scripts/test_e1002_test_report.py
Compliance
- ECSS standards are freely downloadable (ESA); cite the source and
paraphrase per standards-map.yaml.
- compliance: STANDARDS-REF, gated: false.
1---2name: e1002-test-report3description: "Use when produce or review a verification test report against the ECSS-E-ST-10-02 clause 5.3.2.1 and Annex C Document Requirements Definition: confirm the report identifies the article, its configuration, the facility and the test procedure by revision, confirm the cited revision is the one actually run, confirm every test objective carries its own verdict, confirm each observed discrepancy is linked to a raised nonconformance, and aggregate the objective verdicts under a severity ordering. Trigger: ecss, e-st-10-02c, test-report, annex-c-drd, test-objectives, procedure-revision, discrepancy, nonconformance-link, measured-data."4license: Apache-2.05compliance: STANDARDS-REF6standards:7 - id: ecss8 reference-only: true9gated: false10domain: space-systems11pack: space-systems12compatibility: "agentskills.io SKILL.md; any SKILL.md host (Claude Code, Hermes, OpenClaw)"13metadata:14 source: https://skillmd.com/skills/ashfordeou/e1002-test-report15 domain: space-systems16 subdomain: ecss17 tags: [ecss, e-st-10-02c, test-report, annex-c-drd, test-objectives, procedure-revision, discrepancy]18 version: 0.1.019 author: Aero Agent Skills20---2122# ECSS Verification — Test Report DRD (space-systems/ecss/e1002-test-report)2324Use when the task is to produce or check the test report of25ECSS-E-ST-10-02 clause 5.3.2.1 and Annex C -- the record that turns a26test run into verification evidence against a specific article, a27specific facility and a specific procedure revision.2829## Domain quick reference3031- The report cites the test procedure *by revision*, and the cited32 revision must be the one actually executed. This is not a33 typographic class of defect: with the wrong revision the report34 describes a test that was not run, and every conclusion in it is35 attributed to the wrong method.36- An absent revision and a mismatched revision are handled by different37 checks, so a blank field is reported once as missing identification38 rather than twice.39- Identification is what lets data be attributed at all: article,40 configuration, facility, procedure and revision. Data that cannot be41 attributed to a specific article in a specific state is not evidence42 about anything.43- Every test objective carries its own verdict. A report is not a pass44 because most objectives passed -- each objective was written because45 it had to be demonstrated.46- An objective marked not performed needs a reason. Without one, the47 reader cannot distinguish a deliberate descope from an omission, and48 those close very differently.49- Overall verdict is severity ordered: any failure fails the report;50 otherwise anything not performed leaves it incomplete; only an51 all-pass set passes.52- A discrepancy with no nonconformance raised against it is an53 observation nobody is obliged to close. Linking it is what puts it54 into a process that ends in a disposition.55- Summarized results need a reference to the recorded data. Re-analysis56 is what makes test data evidence rather than testimony.5758## Workflow59601. Check the identification fields; report each that is absent.612. Compare the cited procedure revision against the revision actually62 run and report a mismatch.633. For each objective, confirm a verdict is recorded, and that a64 not-performed objective carries its reason; reject duplicate65 objective identifiers.664. For each discrepancy, confirm a nonconformance reference is present.675. Confirm the measured data reference is present.686. Aggregate the objective verdicts under the severity ordering.697. The report is acceptable only when it passes and no finding stands.7071## Pitfalls7273- Citing the procedure without its revision, or carrying the revision74 from the plan rather than from what the operators actually ran.75- Reporting an overall pass because the failing objective was "minor".76 Severity belongs in the disposition of the nonconformance, not in the77 arithmetic of the verdict.78- Marking an objective not performed and leaving the reason to the79 reader, who cannot tell a descope from a miss.80- Recording a discrepancy in narrative text with no nonconformance81 raised. It reads as disclosed and enters no process.82- Presenting summarized results with no pointer to the recorded data,83 which makes independent re-analysis impossible.84- Letting two objectives share an identifier, so one verdict silently85 overwrites another in any downstream roll-up.8687## Behavior contract (gate 3)8889The identification, procedure-revision, objective-verdict,90discrepancy-linkage, data-reference and severity-ordered aggregation91logic is exercised by the gate 3 contract test:92scripts/test_e1002_test_report.py against93scripts/e1002_test_report_logic.py (stdlib unittest, offline). Run:94python3 scripts/test_e1002_test_report.py9596## Compliance9798- ECSS standards are freely downloadable (ESA); cite the source and99 paraphrase per standards-map.yaml.100- compliance: STANDARDS-REF, gated: false.