You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
✅ Redaction System: 266/266 workflows have redact_secrets steps
✅ Token Cascades: 955 instances of GH_AW_GITHUB_MCP_SERVER_TOKEN || secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN fallback pattern
✅ Permission Blocks: 266/266 workflows define explicit permissions:
✅ No secrets in job/step outputs — the raw grep matched 33 lines, but manual inspection confirmed all matches are env: blocks (e.g. ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}) picked up by context overlap with an unrelated outputs: key, not real output exposure.
✅ No template injection risk — occurrences of github.event. are confined to env:-mediated variable assignment (the standard gh-aw safe pattern), not direct inline expression interpolation in run: scripts.
🎯 Key Findings
GITHUB_TOKEN and GH_AW_GITHUB_TOKEN dominate usage (4,397 + 3,764 references, ~93% of all secret refs), consistent with the token-cascade design used for GitHub API/MCP authentication across nearly every workflow.
LLM provider keys are well-isolated: ANTHROPIC_API_KEY (278), OPENAI_API_KEY (105), CODEX_API_KEY (104) — each scoped to specific engine steps, no cross-contamination observed.
Observability secrets are present but bounded: OTEL/Sentry/Grafana auth+endpoint secrets (699+467+465+233 refs) appear only in telemetry-related steps, not exposed elsewhere.
100% redaction and permission-block coverage across all 266 lock files indicates consistent application of the compiler's security scaffolding — no workflow bypasses these controls.
💡 Recommendations
Maintain current cascade pattern: Continue using the GH_AW_GITHUB_MCP_SERVER_TOKEN || secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN fallback for new workflows to preserve least-privilege token scoping.
Monitor growth of GH_AW_SIDE_REPO_PAT and GH_AW_AGENT_TOKEN (24 and 15 refs respectively) — these are broader-scope PATs; ensure they remain limited to workflows that genuinely require cross-repo access.
No immediate action needed — posture is stable with full redaction/permission coverage; continue running this daily scan to catch regressions early.
🔑 Top 10 Secrets by Usage
Rank
Secret Name
Occurrences
Type
1
GITHUB_TOKEN
4,397
GitHub Token
2
GH_AW_GITHUB_TOKEN
3,764
GitHub Token
3
GH_AW_GITHUB_MCP_SERVER_TOKEN
1,752
GitHub Token
4
COPILOT_GITHUB_TOKEN
761
GitHub Token
5
GH_AW_OTEL_SENTRY_AUTHORIZATION
699
Observability
6
GH_AW_OTEL_SENTRY_ENDPOINT
467
Observability
7
GH_AW_OTEL_GRAFANA_AUTHORIZATION
465
Observability
8
ANTHROPIC_API_KEY
278
LLM Provider
9
GH_AW_OTEL_GRAFANA_ENDPOINT
233
Observability
10
OPENAI_API_KEY
105
LLM Provider
📈 Trends
No prior daily-secrets snapshot was found to diff against — this establishes the baseline for future comparisons. Baseline stats:
Total workflows: 266
Total secret refs: 8,734
Token refs: 1,963
Unique secret types: 40
Redaction count: 266
Cascade count: 955
📖 Reference Documentation
For detailed information about secret usage patterns, see:
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
Uh oh!
There was an error while loading. Please reload this page.
🔐 Daily Secrets Analysis Report
Date: 2026-07-29
Workflow Files Analyzed: 266
Run: §30477743812
📊 Executive Summary
secrets.*)github.token)🛡️ Security Posture
✅ Redaction System: 266/266 workflows have
redact_secretssteps✅ Token Cascades: 955 instances of
GH_AW_GITHUB_MCP_SERVER_TOKEN || secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKENfallback pattern✅ Permission Blocks: 266/266 workflows define explicit
permissions:✅ No secrets in job/step outputs — the raw grep matched 33 lines, but manual inspection confirmed all matches are
env:blocks (e.g.ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}) picked up by context overlap with an unrelatedoutputs:key, not real output exposure.✅ No template injection risk — occurrences of
github.event.are confined toenv:-mediated variable assignment (the standard gh-aw safe pattern), not direct inline expression interpolation inrun:scripts.🎯 Key Findings
GITHUB_TOKENandGH_AW_GITHUB_TOKENdominate usage (4,397 + 3,764 references, ~93% of all secret refs), consistent with the token-cascade design used for GitHub API/MCP authentication across nearly every workflow.ANTHROPIC_API_KEY(278),OPENAI_API_KEY(105),CODEX_API_KEY(104) — each scoped to specific engine steps, no cross-contamination observed.💡 Recommendations
GH_AW_GITHUB_MCP_SERVER_TOKEN || secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKENfallback for new workflows to preserve least-privilege token scoping.GH_AW_SIDE_REPO_PATandGH_AW_AGENT_TOKEN(24 and 15 refs respectively) — these are broader-scope PATs; ensure they remain limited to workflows that genuinely require cross-repo access.🔑 Top 10 Secrets by Usage
📈 Trends
No prior daily-secrets snapshot was found to diff against — this establishes the baseline for future comparisons. Baseline stats:
📖 Reference Documentation
For detailed information about secret usage patterns, see:
scratchpad/secrets-yml.mdactions/setup/js/redact_secrets.cjsGenerated: 2026-07-29T18:00 UTC
Workflow: Daily Secrets Analysis Agent
All reactions