aliasunder/umm-actually
LLM PR review via OpenRouter — one consolidated review, inline findings
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| github_token | Token used to fetch the diff and submit the review (a GitHub App installation token keeps the bot identity) | yes | — |
| openrouter_api_key | OpenRouter API key | yes | — |
| model | OpenRouter model slug exactly as listed on openrouter.ai/models, e.g. anthropic/claude-sonnet-4-6 | no | anthropic/claude-sonnet-4-6 |
| fallback_model | Model to retry with if the primary model fails the structured-output ladder | no | "" |
| max_findings | Cap on posted findings, highest severity first. Empty = uncapped (all validated findings post) | no | "" |
| severity_threshold | Minimum severity to post: low | medium | high | critical | no | low |
| conventions_file | Repo-relative path to the conventions file included in the prompt | no | AGENTS.md |
| phases | Review phases to run. V1 supports: combined | no | combined |
| context_budget_tokens | Approximate token budget for prompt context (file contents + diff — conventions have a separate cap) | no | 80000 |
| trace_related_files | Include related files in the prompt — import-tracing for caller regressions, and doc-mention scanning for staleness detection | no | true |
| priority_docs | Comma-separated repo-relative paths always included in review context (empty = disabled) | no | README.md |
| max_scan_files | Maximum files to walk during workspace scan for related file and doc detection | no | 5000 |
| max_scan_bytes | Maximum byte size of a single file to include in the workspace scan | no | 262144 |
| max_related_files | Maximum import-traced related files to include in review context | no | 8 |
| max_related_docs | Maximum mention-matched documentation files to include in review context (excludes priority docs) | no | 4 |
| cost_summary | Write a per-run cost report (model, prompt/completion tokens, USD) to the workflow step summary | no | true |
| pr_number | PR number override — required only when the triggering event does not identify a PR directly | no | "" |
Outputs
| name | description |
|---|---|
| findings_count | Number of new findings successfully posted this run (after threshold, cap, and cross-run dedup; findings whose post failed are excluded and re-report on the next run) |
| review_url | URL of the submitted review; empty when no review was posted |
| model_used | Model that produced the accepted response |
| skipped_reason | Non-empty when the review was skipped (e.g. diff too large) |