[self-care:open-source-failures] Digest of clustered open-source workflow failures from 2026-09-04 snapshot

Bounded seven-day snapshot across 7 public projects found 665 failed runs in the newest 100 available failure records, which cluster into 2 P0, 4 P1, and 15 P2 findings with no existing `[self-care:open-source-failures]` coverage found in `githubnext/gh-aw-cao` open issues. The highest-priority blockers are repeated live-mode target-authority precompute failures across AW Doctor and Dependabot workflows, and repeated GitHub App token generation failures in AI Credit Savings.

**Action:** CAO maintainers should triage the two remediation issues below first; accept when live-mode self-care and dependabot runs stop failing for the cited signatures in a fresh snapshot.

Critical prioritized remediation targets:
- `githubnext/gh-aw-cao` — repeated `Target authority missing: add .github/workflows/cao.json to the target default branch for live mode` across AW Doctor and Dependabot workflows (P0, 48 corroborating runs across 4 workflows in the bounded snapshot).
- `githubnext/gh-aw-cao` — repeated `Generate GitHub App token` failures in `optimization-ai-credit-optimizer.lock.yml` (P1, 8 runs).
- `githubnext/gh-aw-cao` — repeated `Process Safe Outputs` failures in `software-development-practices-github-well-architected.lock.yml` (P1, 13 runs), not filed this round because issue budget is capped at two remediation issues.

Snapshot summary

- Snapshot generated at: `2026-09-04T23:04:50.067Z`
- Evidence window: 168 hours
- Public projects scanned: 7
- Failed runs in snapshot metadata: 665
- Available bounded failure records analyzed: newest 100 only (`truncated: true`)
- Existing open `[self-care:open-source-failures]` issues found: 0

Severity counts from the bounded evidence:
- P0: 2
- P1: 4
- P2 / insufficient evidence: 15

Representative untracked clusters from the bounded evidence:
1. `githubnext/gh-aw-cao` / AW Doctor + Dependabot workflows — target-authority precompute failures (`pre_activation` / `Run CAO control precompute`) with explicit messages about missing or invalid `.github/workflows/cao.json` target authority.
2. `githubnext/gh-aw-cao` / `optimization-ai-credit-optimizer.lock.yml` — repeated `Generate GitHub App token` failures.
3. `githubnext/gh-aw-cao` / `software-development-practices-github-well-architected.lock.yml` — repeated `Process Safe Outputs` failures.
4. `github/gh-aw-mcpg` / `readonly-stress-gvisor.lock.yml` — two failures at `Execute GitHub Copilot CLI`.

### Control Plane
- Correlation ID: `33928821155-188`
- Central repository: `githubnext/gh-aw-cao`
- Control-plane run: https://github.com/githubnext/gh-aw-cao/actions/runs/33928821155

> Generated by [SelfCare / Open Source Failures](https://github.com/githubnext/gh-aw-cao/actions/runs/33929088752) · pi · gpt54 · 25.9 AIC · ⌖ 8.51 AIC · ⊞ 8.3K · [◷](https://github.com/search?q=repo%3Agithubnext%2Fgh-aw-cao+is%3Aissue+%22gh-aw-workflow-call-id%3A+githubnext%2Fgh-aw-cao%2Fself-care-open-source-failures%22&type=issues)
> - [x] expires on Sep 18, 2026, 11:26 PM UTC

Start with the cited AW Doctor and Dependabot failures, then inspect .github/workflows/cao.json and the precompute step for the target-authority error. Review optimization-ai-credit-optimizer.lock.yml for the GitHub App token failures and compare both against a fresh seven-day snapshot. Done means the cited live-mode and Dependabot signatures no longer recur.

Written by the indexing model from the issue text.