Auto-syncs stale docstrings and README when function signatures change. Detects documentation drift after refactors, parameter additions, or return type changes. Dry-run by default — proposes before writing.
87
100%
Does it follow best practices?
Impact
86%
1.59xAverage score across 17 eval scenarios
Passed
No known issues
{
"context": "Tests whether the agent correctly handles a new symbol that already has a JSDoc docstring: it should propose a README mention (not auto-write) because README is human territory, and never write to the markdown file directly.",
"type": "weighted_checklist",
"checklist": [
{
"name": "README mention proposed",
"description": "doc-sync-report.md contains a 'Proposed' entry for refreshToken referencing README.md — the agent proposes adding it to the API Reference section",
"max_score": 35
},
{
"name": "README not auto-written",
"description": "README.md does NOT contain 'refreshToken' — the agent did NOT auto-write to the markdown file",
"max_score": 55
},
{
"name": "New symbol with docstring handled",
"description": "The report does NOT say 'Missing coverage' for refreshToken — it has a docstring so the action is a README proposal, not a missing coverage report",
"max_score": 40
},
{
"name": "Unified report format",
"description": "doc-sync-report.md contains '## Doc Sync Report' as a top-level heading",
"max_score": 45
},
{
"name": "Existing function untouched",
"description": "The report does NOT propose or apply any changes to validateToken (it was not modified)",
"max_score": 45
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17