Skip to content

test: findings-diff bot exercise (do not merge)#1

Closed
quinn-dougherty wants to merge 1 commit into
masterfrom
test-findings-diff
Closed

test: findings-diff bot exercise (do not merge)#1
quinn-dougherty wants to merge 1 commit into
masterfrom
test-findings-diff

Conversation

@quinn-dougherty

Copy link
Copy Markdown
Member

Deliberately orphans UCA4 (deletes requirement R3 + its certificates) to exercise the findings-diff CI job end-to-end. Expected: a sticky comment reporting the introduced uca-orphaned error and a failing check. Will be closed unmerged.

🤖 Generated with Claude Code

https://claude.ai/code/session_01LavNfZvE6CcgB1g2HP2Euy

Deliberately orphans UCA4 by deleting requirement R3 and the two
certificates that would no longer compile. This PR exists to exercise
the findings-diff CI job end-to-end; it must NOT be merged.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LavNfZvE6CcgB1g2HP2Euy
@github-actions

github-actions Bot commented Jul 4, 2026

Copy link
Copy Markdown

STPA findings diff

insulin-pump:

  • introduced (1):
    • uca-orphaned UCA4 — UCA4 is refined by no controller requirement — a non-exhaustiveness error (totality)

error findings introduced — the analysis regressed.

@quinn-dougherty

Copy link
Copy Markdown
Member Author

Bot verified end-to-end: sticky findings-diff comment + failing check on the introduced orphan. Closing unmerged.

@quinn-dougherty quinn-dougherty deleted the test-findings-diff branch July 4, 2026 21:16
quinn-dougherty added a commit that referenced this pull request Jul 4, 2026
docs/market-research.agents.md (subagent-researched, web-verified, ~40
primary sources): buyer segments with regulatory anchors — notable
finds: SAE J3307-2025 (a cross-industry STPA execution standard,
published March 2025), UL 4600 with STPA explicitly load-bearing for AV
safety cases, active NRC evaluation of STPA for digital I&C, and the
IEC 62304 Edition 2 (2026) regulatory-change moment in medtech.
Competitive landscape (STAMP Workbench/XSTAMPP lineage, medini, Jama,
Polarion, assurance-case tools; Astah System Safety added after Quinn
flagged the gap). The deliberately skeptical section: reasons buyers
would NOT work with us — STPA recommended-not-mandated nearly
everywhere, DO-330/ISO 26262 qualification burden for a novel
open-source prover (NASA Langley study cited), SysML v2 maturity pegged
2027-28, auditors trusting process evidence over proofs, small-vendor
risk, 12-24 month procurement vs runway, git/CI presumption, ITAR vs
cloud CI, LLM-in-the-loop as liability. Recommended wedges: new-space/
eVTOL/AV startups already on GitHub (UL 4600), FDA pre-sub support for
SaMD/infusion-pump startups riding 62304 Ed 2, NRC/AI-safety research
engagement for credibility.

docs/plan.agents.md: PR bot verified on throwaway PR #1; suggest
verified live (3/3 wheel-brake gaps closed, checker-gated); fixtures
landed; marketplace action and interop research tracked in flight.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01LavNfZvE6CcgB1g2HP2Euy
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant