📊 Agentic Workflow Lock File Statistics - February 8, 2026 #14484
Closed
Replies: 1 comment
-
|
This discussion was automatically closed because it expired on 2026-02-15T08:28:12.160Z.
|
Beta Was this translation helpful? Give feedback.
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
Executive Summary
This comprehensive analysis examines all
.lock.ymlfiles in the.github/workflows/directory to identify usage patterns, popular triggers, safe outputs, structural characteristics, and trends over time.File Size Distribution
Key Statistics:
View Smallest and Largest Workflows
5 Smallest Workflows:
5 Largest Workflows:
Trigger Analysis
Most Popular Triggers
workflow_dispatchschedulepull_requestissuesdiscussion_commentdiscussionpushKey Insight: 88.4% of workflows support manual triggering via
workflow_dispatch, making the system highly interactive and testable.Common Trigger Combinations
schedule+workflow_dispatchworkflow_dispatchonlyschedule+workflow_dispatch+pull_requestissuesonlyworkflow_dispatch+issuesPattern: The most common pattern (66.7%) combines scheduled automation with manual override capability.
View Schedule Patterns (Cron)
Most Common Schedule Times (all weekday-only patterns):
0 14 * * 1-50 13 * * 1-50 11 * * 1-50 9 * * 1-50 7 * * 1-5Pattern: Schedules are intentionally scattered throughout the day (7 AM - 4 PM UTC) to distribute load. Most workflows run on weekdays only (Monday-Friday).
Interesting Schedules:
*/30 * * * *- Every 30 minutes (1 workflow)59 */6 * * *- Every 6 hours at :59 minutes (1 workflow)0 9 * * 2,4- Tuesday and Thursday only (1 workflow)Safe Outputs Analysis
Safe outputs are the primary mechanism for workflows to produce visible results (issues, discussions, comments, etc.).
Safe Output Types Distribution
noopmissing_toolmissing_dataadd-commentKey Finding: All workflows use the "safeoutputs" pattern with
noop,missing_tool, andmissing_datatools for reporting. Only 17% use theadd-commentcapability for interactive feedback.Notable: While the analysis detected safe output tool availability, only 25 workflows (17%) actively use the commenting capability, suggesting most workflows prefer creating discussions or issues over commenting on existing items.
Structural Characteristics
Job Complexity
Average Workflow Structure:
Distribution:
Most Common: 7-8 jobs per workflow (57.8% of all workflows)
View Most and Least Complex Workflows
Most Complex (by job count):
Least Complex:
By step count (most steps):
Typical Lock File Structure
Based on statistical analysis, a typical
.lock.ymlfile has:schedule+workflow_dispatch(66.7% pattern)contents: read,discussions: write,issues: write,pull-requests: writePermission Patterns
Most Common Permissions
contents: readdiscussions: writeissues: writepull-requests: writecontents: writeSecurity Pattern: All workflows request minimal
contents: readpermission. Write access tocontentsis granted to only 42% of workflows, showing a principle of least privilege.Key Insight: 95% of workflows have write permissions for discussions, issues, and pull requests, indicating the primary output mechanism is GitHub's collaboration features rather than code changes.
Tool & MCP Patterns
Most Used MCP Servers
githubplaywrightNote: The
githubMCP server is heavily used across all workflows (avg ~12.6 mentions per workflow), indicating extensive GitHub API integration.Concurrency Patterns
group: "gh-aw-${{ github.workflow }}"Insight: Every workflow uses concurrency control to prevent simultaneous runs of the same workflow, ensuring resource efficiency and preventing conflicts.
Timeout Configuration
Timeout Distribution
Average Timeout: ~16.5 minutes per job
Pattern: 89.5% of jobs use 10-20 minute timeouts, indicating most agentic tasks complete within this window.
Interesting Findings
High Interactivity: 88% of workflows support manual triggering, making the system highly testable and interactive.
Weekday-Focused Automation: Scheduled workflows predominantly run on weekdays (Monday-Friday), with times scattered between 7 AM - 4 PM UTC to distribute load.
Consistent Job Structure: 58% of workflows have 7-8 jobs, suggesting a standardized workflow template or pattern (likely: pre_activation, activation, agent, collect_output, and engine-specific jobs).
Multi-Trigger Complexity: Only 5 workflows have 4+ trigger types, with
scout.lock.yml,q.lock.yml,pr-nitpick-reviewer.lock.yml, andcloclo.lock.ymlbeing the most versatile.Smoke Test Outliers: The 3 largest workflows (90+ KB) are smoke tests (
smoke-claude,smoke-copilot,smoke-opencode), indicating comprehensive testing suites.Universal Concurrency: 100% adoption of concurrency groups shows mature workflow design preventing race conditions.
Low Write Access: Only 42% of workflows have
contents: writepermission, but 95% can write to discussions/issues, showing a preference for communication over code modification.Historical Trends
Comparing with previous analysis from 2026-02-07:
Key Trend: Workflows have become more efficient - total size decreased by 5.7% while maintaining the same number of files and job structure. The decrease in total steps (18.7%) suggests optimization or removal of redundant steps.
Observation: The largest file size decreased from 109 KB to 99 KB, indicating recent optimizations to the most complex workflows.
Recommendations
Template Standardization: With 58% of workflows using 7-8 jobs, consider documenting this as the "standard workflow pattern" to help new workflow authors.
Timeout Optimization: 90% of jobs use 10-20 minute timeouts. Consider analyzing actual runtime to see if timeouts can be reduced further for faster failure detection.
Load Distribution: Schedule patterns show good time distribution (7 AM - 4 PM UTC). Continue this pattern to avoid CI/CD resource contention.
Size Monitoring: Track the size trend over time. The recent 5.7% decrease is positive, but ensure critical functionality isn't being removed.
Trigger Usage: Only 9.5% of workflows use
pull_requesttriggers. Consider expanding automated PR workflows if code review automation would be beneficial.Permission Review: The 42% of workflows with
contents: writeshould be audited to ensure they truly need write access.Methodology
.github/workflows/*.lock.ymlReferences:
Analysis performed on 2026-02-08 08:24:05 UTC
Beta Was this translation helpful? Give feedback.
All reactions