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
✅ No direct github.event.* interpolation found in run: scripts (TestCompiledLockFiles_NoGitHubEventExpressionsInRunScripts passed)
✅ No secrets detected in job outputs (TestCompiledLockFiles_NoSecretsInOutputs passed)
🎯 Key Findings
GITHUB_TOKEN dominates usage — 5,348 occurrences, consistent with its role as the default auth token across nearly every job/step.
Token cascade pattern is pervasive — GH_AW_GITHUB_MCP_SERVER_TOKEN (1,945 refs) combined with 1,014 full three-way fallback chains shows the MCP server auth design is applied uniformly, not just in a handful of workflows.
Observability secrets are significant — Sentry (744+497 for auth/endpoint) and Grafana (495+248) OTEL credentials account for ~1,984 references, reflecting broad telemetry instrumentation across workflows.
AI provider keys are diverse but low-volume individually — OPENAI_API_KEY (463), CODEX_API_KEY (462), ANTHROPIC_API_KEY (255), plus smaller counts for GEMINI, KIRO, CURSOR, FOUNDRY — indicating multi-engine support with per-engine credential isolation rather than a shared universal key.
Long tail of specialized integration secrets — Datadog (DD_API_KEY, DD_APP_KEY, etc.), Jira, Linear, Slack, and Squad GitHub App credentials each appear only 1–8 times, consistent with narrowly-scoped integration workflows.
💡 Recommendations
Maintain redaction parity: Redaction coverage is currently 100% (299/299) — ensure any newly added workflow templates include the redaction step by default in CI validation.
Audit low-usage secrets periodically: Secrets referenced only 1–2 times (e.g. GH_AW_OTEL_DATADOG_ENDPOINT, JIRA_API_TOKEN, SLACK_BOT_TOKEN) should be reviewed quarterly to confirm the workflows using them are still active and the secrets aren't orphaned.
Continue enforcing template-injection and output-exposure tests: Both TestCompiledLockFiles_NoGitHubEventExpressionsInRunScripts and TestCompiledLockFiles_NoSecretsInOutputs passed cleanly; keep these as required CI gates for every compiled lock file.
🔑 Top 15 Secrets by Usage
Rank
Secret Name
Occurrences
Type
1
GITHUB_TOKEN
5348
GitHub Token
2
GH_AW_GITHUB_TOKEN
4544
GitHub Token
3
GH_AW_GITHUB_MCP_SERVER_TOKEN
1945
GitHub Token
4
GH_AW_OTEL_SENTRY_AUTHORIZATION
744
Observability
5
COPILOT_GITHUB_TOKEN
575
GitHub Token
6
GH_AW_OTEL_SENTRY_ENDPOINT
497
Observability
7
GH_AW_OTEL_GRAFANA_AUTHORIZATION
495
Observability
8
OPENAI_API_KEY
463
AI Provider
9
CODEX_API_KEY
462
AI Provider
10
ANTHROPIC_API_KEY
255
AI Provider
11
GH_AW_OTEL_GRAFANA_ENDPOINT
248
Observability
12
GH_AW_DEFAULT_OTLP_HEADERS
102
Observability
13
GH_AW_CI_TRIGGER_TOKEN
70
GitHub Token
14
GH_AW_SIDE_REPO_PAT
24
GitHub Token
15
GH_AW_AGENT_TOKEN
15
GitHub Token
📈 Trends
No prior-day stats file was available in this run's ephemeral environment to diff against, since /tmp/gh-aw/agent/secrets-stats.json does not persist between separate workflow runs. Compare this report's raw numbers against the previous "[daily secrets]" discussion (#58840, 2026-09-05) manually if a longitudinal trend is needed. This run's snapshot:
Total workflows: 299
Total secrets.* references: 10,370
Total github.token references: 2,159
Unique secret types: 41
Redaction coverage: 299/299
Token cascade count: 1,014
📖 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: 2026-09-06
Workflow Files Analyzed: 299
Run: https://github.com/github/gh-aw/actions/runs/34049222194
📊 Executive Summary
secrets.*)github.token)secrets.*matches occur inside step blocks)🛡️ Security Posture
✅ Redaction System: 299/299 workflows have redaction steps (100% coverage)
✅ Token Cascades: 1,014 instances of
GH_AW_GITHUB_MCP_SERVER_TOKEN || secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKENfallback chains✅ Permission Blocks: 299 explicit
permissions:definitions (one per workflow, least-privilege enforced)✅ No direct
github.event.*interpolation found inrun:scripts (TestCompiledLockFiles_NoGitHubEventExpressionsInRunScriptspassed)✅ No secrets detected in job outputs (
TestCompiledLockFiles_NoSecretsInOutputspassed)🎯 Key Findings
GH_AW_GITHUB_MCP_SERVER_TOKEN(1,945 refs) combined with 1,014 full three-way fallback chains shows the MCP server auth design is applied uniformly, not just in a handful of workflows.💡 Recommendations
GH_AW_OTEL_DATADOG_ENDPOINT,JIRA_API_TOKEN,SLACK_BOT_TOKEN) should be reviewed quarterly to confirm the workflows using them are still active and the secrets aren't orphaned.TestCompiledLockFiles_NoGitHubEventExpressionsInRunScriptsandTestCompiledLockFiles_NoSecretsInOutputspassed cleanly; keep these as required CI gates for every compiled lock file.🔑 Top 15 Secrets by Usage
📈 Trends
No prior-day stats file was available in this run's ephemeral environment to diff against, since
/tmp/gh-aw/agent/secrets-stats.jsondoes not persist between separate workflow runs. Compare this report's raw numbers against the previous "[daily secrets]" discussion (#58840, 2026-09-05) manually if a longitudinal trend is needed. This run's snapshot:secrets.*references: 10,370github.tokenreferences: 2,159📖 Reference Documentation
For detailed information about secret usage patterns, see:
scratchpad/secrets-yml.mdactions/setup/js/redact_secrets.cjsGenerated: 2026-09-06 17:42 UTC
Workflow: Daily Secrets Analysis Agent
All reactions