[detection-stats] Detection stats for github/gh-aw - 2026-09-03

745 external-detector runs analysed · 0% detection-job error rate · 72 soft failures · 0% threat rate.

## Summary

# gh-aw detection statistics - 2026-09-03 (UTC)

Repository: `github/gh-aw`
Window: `2026-09-03T00:00:00Z` .. `2026-09-03T23:59:59Z`
API requests: 1942, rate-limit pauses: 1
Data complete: yes

## Totals

| Metric | Count |
|---|---|
| Workflow runs in window | 1981 |
| Agentic runs (`*.lock.yml`) | 1185 |
| Runs with a `detection` job | 784 |
| ... using the external detector | 745 |
| ... using the built-in detector | 39 |
| ... detector could not be determined | 0 |
| Agentic runs without a `detection` job | 401 |

All rates below are over the **external detector** population (745 runs). Since gh-aw #54111 the external detector is the compile-time default; a run counts as external when its `detection` job showed the `Install threat-detect binary` step, when another run of the same workflow did that day, or when the workflow is `.lock.yml` and no run showed the built-in shape (a completed detection job with steps but no marker). Runs on a workflow that opted out with `features: gh-aw-detection: false` count as built-in.

## Detection job outcomes

| Outcome | Count | Rate |
|---|---|---|
| `success` | 401 | 53.83% |
| `skipped` | 343 | 46.04% |
| `cancelled` | 1 | 0.13% |

**Error rate (failure/timed_out/action_required): 0%**

## Verdict availability

| State | Meaning | Count |
|---|---|---|
| `present` | detection artifact downloaded and parsed | 294 |
| `absent` | detection job ran but published no artifact (soft failure) | 72 |
| `skipped` | detection job was skipped or was still running (nothing to fetch) | 343 |
| `unreadable` | artifact zip could not be unpacked | 36 |

Green detection jobs that published no verdict: **72** (detection steps are `continue-on-error`, so a missing verdict artifact is the only reliable signal for these).

## Detection results

| Result | Count |
|---|---|
| Runs with a parsed verdict | 294 |
| Clean (no threat) | 294 |
| Any threat | 0 |
| `prompt_injection` | 0 |
| `secret_leak` | 0 |
| `malicious_patch` | 0 |

**Threat rate (of runs with a verdict): 0%**

## By workflow (top entries)

| Workflow | Runs | Failed | Cancelled | Skipped | No verdict | Threats |
|---|---|---|---|---|---|---|
| Q | 74 | 0 | 0 | 74 | 0 | 0 |
| PR Sous Chef | 69 | 0 | 0 | 0 | 0 | 0 |
| Deployment Incident Monitor | 66 | 0 | 0 | 65 | 0 | 0 |
| Daily Trajectory Grader Implementer | 43 | 0 | 0 | 43 | 0 | 0 |
| [aw] Failure Investigator (6h) | 43 | 0 | 0 | 38 | 0 | 0 |
| Issue Monster | 40 | 0 | 0 | 5 | 4 | 0 |
| Avenger | 22 | 0 | 0 | 9 | 13 | 0 |
| Impeccable Skills Reviewer | 16 | 0 | 0 | 0 | 1 | 0 |
| Matt Pocock Skills Reviewer | 16 | 0 | 0 | 0 | 0 | 0 |
| PR Code Quality Reviewer | 16 | 0 | 0 | 0 | 0 | 0 |
| Ponytail Reviewer | 16 | 0 | 0 | 0 | 4 | 0 |
| Test Quality Sentinel | 16 | 0 | 0 | 0 | 2 | 0 |
| Design Decision Gate 🏗️ | 15 | 0 | 0 | 0 | 0 | 0 |
| Auto-Triage Issues | 14 | 0 | 0 | 4 | 1 | 0 |
| Daily Go Test Parallelizer | 12 | 0 | 0 | 0 | 2 | 0 |
| Workflow Generator | 10 | 0 | 0 | 10 | 0 | 0 |
| Contribution Check | 6 | 0 | 0 | 0 | 0 | 0 |
| Code Scanning Fixer | 4 | 0 | 0 | 2 | 0 | 0 |
| Deep Report | 4 | 0 | 0 | 0 | 1 | 0 |
| PR Triage Agent | 4 | 0 | 0 | 0 | 2 | 0 |
| Daily Windows Defender Release Scan | 3 | 0 | 0 | 3 | 0 | 0 |
| Daily Credit Limit Test | 2 | 0 | 0 | 1 | 0 | 0 |

_219 further workflows omitted; see the detection-stats artifact for the full list._

_48 further notable runs omitted; see the detection-stats artifact._

## Change since 2026-09-02

| Metric | 2026-09-02 | 2026-09-03 | Delta | 7-day mean |
|---|---|---|---|---|
| External detector runs | 998 | 745 | -253 | 725.7 |
| Error rate (%) | 0.0 | 0.0 | 0.0 | 0.0 |
| Soft failures | 76 | 72 | -4 | 56.7 |
| Runs with verdict | 282 | 294 | +12 | n/a |
| Any threat | 1 | 0 | -1 | n/a |
| Threat rate (%) | 0.35 | 0.0 | -0.35 | 0.21 |

## Watch list

- Avenger — 0 failed, 13 without a verdict, out of 22 runs
- Issue Monster — 0 failed, 4 without a verdict, out of 40 runs
- Ponytail Reviewer — 0 failed, 4 without a verdict, out of 16 runs
- Test Quality Sentinel — 0 failed, 2 without a verdict, out of 16 runs
- Daily Go Test Parallelizer — 0 failed, 2 without a verdict, out of 12 runs

Collected by: https://github.com/github/gh-aw-threat-detection/actions/runs/33834956654
Full data: the detection-stats-33834956654 artifact on that run.

> Generated by [Detection Stats Daily](https://github.com/github/gh-aw-threat-detection/actions/runs/33834956654) · copilot · auto · 36.8 AIC · ⌖ 6.68 AIC · ⊞ 11.3K · [◷](https://github.com/search?q=repo%3Agithub%2Fgh-aw-threat-detection+is%3Aissue+%22gh-aw-workflow-call-id%3A+github%2Fgh-aw-threat-detection%2Fdetection-stats-daily%22&type=issues)

Start by reviewing the detection-stats-33834956654 artifact and the linked Detection Stats Daily workflow run. The issue reports daily metrics but does not identify a requested code or documentation change, so the work is not complete until a concrete follow-up is defined.

Written by the indexing model from the issue text.