Repository navigation
[lockfile-stats] Lockfile Statistics Report — 2026-10-11 #67778
Closed
Replies: 1 comment
|
This discussion was automatically closed because it expired on 2026-10-12T20:33:03.408Z.
|
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
337 lockfiles analyzed on 2026-10-11 (0 skipped), up +3 from 334 the prior day. Total size 52.0 MB (avg 158.1 KB/file, min 87.6 KB, max 283.0 KB).
Trigger Analysis
Top trigger combinations:
schedule+workflow_dispatch(208, 61.7%),workflow_dispatchonly (71, 21.1%),pull_request+schedule+workflow_dispatch(32, 9.5%, +3),pull_request+workflow_dispatch(9).Dominant cron cadence:
0 0 */2 * *(every 2 hours) — 53 workflows (15.7% of all lockfiles, +3), far ahead of any other schedule.Safe Outputs Analysis
Core infra safe-outputs (
missing_data,missing_tool,noop,report_incomplete,create_report_incomplete_issue) appear in 330/337 lockfiles (97.9%).Full safe-output type breakdown (58 types)
create_pull_request_review_comment 16, submit_pull_request_review 16, upload_artifact 15, create_check_run 14, update_issue 10, push_to_pull_request_branch 10, max_bot_mentions 9, dispatch_workflow 9, ledger_append 7, close_issue 7, remove_labels 7, ledger_request_compaction 6, comment_memory 6, assign_to_agent 5, link_sub_issue 4, update_pull_request 4, close_pull_request 4, send-slack-message 4, create_code_scanning_alert 3, close_discussion 3, reply_to_pull_request_review_comment 3, set_issue_type 3, and 20+ types at count ≤2 (hide_comment, assign_to_user, create_agent_session, upload_code_coverage, jira_*, linear_create_issue, update_project, etc.)
Discussion categories (93 workflows configure
create_discussion; category resolved for 92/93 — 1 unresolved:pr-sous-chef.lock.yml):safe_outputs_config_missing: 1 lockfile (noGH_AW_SAFE_OUTPUTS_CONFIGfound at all). Self-check passed:create_discussion_workflows(93) > 0 and detection gap (1/93 = 1.1%) is within the 10% threshold, so this is a genuine small gap, not a detection failure.Structural Characteristics
Permission Patterns (agent job)
The agent job itself is read-only on every tracked scope except the write-only
copilot-requestsscope — all writes are isolated to separate safe-output jobs.Union write permissions (any job in the workflow, i.e. the safe-output jobs):
permissions_unknown: 0 — every lockfile's job permissions resolved cleanly.Engine Distribution
(from
gh-aw-metadataagent_id; 0 unresolved)Tool & MCP Patterns
7 lockfiles (2.1%) had no
gh-aw-manifestand were parsed via the legacy# - mcp__...comment fallback.Historical Trends (vs 2026-10-10)
0 0 */2 * *cronInteresting Findings
pull_request+schedule+workflow_dispatchtrigger combo (+3),create_issuesafe output (+3),copilot-requests:writepermission (+3), and the 10/45/60-min timeout trio (+3 each) — consistent with 3 new copilot-engine workflows from a common template.copilot-requestsscope, while 99.4% of workflows (335/337) grant write access somewhere via a dedicated job.0 0 */2 * *(every 2 hours) alone accounts for 15.7% of all lockfiles — far more than any other cron pattern.create_discussioncategories, 79 (85.9%) targetaudits— this very report included.Recommendations
pr-sous-chef.lock.yml(1 of 93create_discussionworkflows) to reach 100% detection.# - mcp__...comment scraping onto thegh-aw-manifestformat.Methodology
single-script compact JSON analysis
References: §38172016542
All reactions