FrancyJGLisboa/agent-skills-platform / references/examples/pr-blocker-summarizer
pr-blocker-summarizer
UnverifiedSummarizes open pull requests into a blockers-first standup digest. Activates when the user asks to summarize open PRs, find blocked pull requests, generate a PR standup, or triage review backlog from a PR export.
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 1.9 KB |
| SKILL.md | skill md | 2.7 KB |
| VERIFICATION.md | markdown | 849 B |
| contracts/data_contract_pull_request_reviews/.contract_id | other | 20 B |
| contracts/data_contract_pull_request_reviews/code/contract_health.py | script | 402 B |
| discovery.json | config | 2.2 KB |
| evals/golden/case-1/expected.json | config | 396 B |
| evals/golden/case-1/input.json | config | 339 B |
| evals/golden/case-2/expected.json | config | 152 B |
| evals/golden/case-2/input.json | config | 209 B |
| evals/golden/case-3/expected.json | config | 233 B |
| evals/golden/case-3/input.json | config | 106 B |
| evals/pr-blocker-summarizer.eval.md | markdown | 1.8 KB |
| scripts/dependency_health.py | script | 3.9 KB |
| scripts/evolve.pyexecutable | script | 10.2 KB |
| scripts/review_staleness.py | script | 6.2 KB |
| scripts/run_evals.pyexecutable | script | 49.4 KB |
| scripts/run_pipeline.pyexecutable | script | 2.7 KB |
| scripts/schema_drift.py | script | 7.3 KB |
| scripts/skill_document.py | script | 9.8 KB |
| scripts/staleness_check.py | script | 7.5 KB |
| scripts/success_ledger.py | script | 16.5 KB |
SKILL.md
View on GitHub---
name: pr-blocker-summarizer
description: Summarizes open pull requests into a blockers-first standup digest. Activates when the user asks to summarize open PRs, find blocked pull requests, generate a PR standup, or triage review backlog from a PR export.
license: MIT
metadata:
author: agent-skills-platform
version: 1.0.0
created: 2026-06-27
last_reviewed: 2026-07-20
review_interval_days: 180
---
# PR Blocker Summarizer
Turn a JSON export of open pull requests into a blockers-first digest: which PRs
are blocked (failing checks, requested changes, or stale), which are ready to
merge, and a one-line count an agent can post to standup.
A bundled **example** skill — small but real, used to demonstrate the creator's
validation, pipeline, and eval-rollout machinery.
## Activation
Activates on "summarize open PRs", "what's blocking my PRs", "PR standup",
"review backlog". Do **not** activate on general git/GitHub questions unrelated to
triaging a set of PRs.
## Input
A JSON array of PRs, each with `title`, `state` (`open`), `checks`
(`passing`/`failing`), `review` (`approved`/`changes_requested`/`pending`), and
`age_days`. Missing fields are treated conservatively (counted as not-blocked only
when clearly ready).
## Run
```bash
python3 scripts/run_pipeline.py --input prs.json --output digest.json
```
Output JSON shape:
```json
{
"total": 7,
"blocked": [{"title": "...", "reasons": ["failing checks"]}],
"ready": ["..."],
"summary": "7 open · 3 blocked · 2 ready to merge"
}
```
## Gotchas
- **`blocked` + `ready` does not equal `total`.** A PR is `ready` only when
`checks` is explicitly `passing` *and* `review` is explicitly `approved`. A PR
with no blockers but a missing `checks` field lands in neither list and vanishes
from both counts. Never narrate the summary line as if the two numbers account
for every PR — say "N blocked, M ready" and leave the remainder unclaimed.
- **`state` is read from the input but never filtered on.** A closed or merged PR
left in the export is counted in `total` and classified like any open one. Filter
the export before running if it may contain non-open PRs.
- **`review: pending` counts as blocked** ("awaiting review"). A PR opened five
minutes ago is reported as a blocker. This is intentional for standup triage, but
it inflates the blocked count on teams that open PRs early.
- **A non-numeric `age_days` silently skips the stale check** rather than erroring.
An export with `"age_days": "unknown"` will never flag anything stale, and the
digest gives no sign that the check did not run.
## Anti-goals
- Does not call the GitHub API; it works on an exported PR list.
- Does not merge or comment on PRs; it only summarizes.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| add-product-e2e-evalpaperclipai/paperclipAdd or extend a Paperclip full-stack runner E2E workflow, fixture, matcher, or report evidence path for local or Daytona execution. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| add-runner-evalpaperclipai/paperclipAdd or extend a Paperclip Runner protocol evaluation definition, roster, assertion, or report fixture with provenance and narrow validation. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| diagnose-why-work-stoppedpaperclipai/paperclipCompatibility entry point for failed, stalled, looping, or over-recovered Paperclip work. Use fix-failed-paperclip-run for the full diagnosis, regression, repair, real verification, and green-PR workflow; honor an explicitly requested diagnosis-only scope. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |
| fix-failed-paperclip-runpaperclipai/paperclipDiagnose and fix failed, stuck, slow, or incorrectly blocked Paperclip tasks and runs. Use with a task/run URL or failure report to trace the root cause, prove a regression red then green, check second-order effects, verify the real workflow, and deliver a PR with passing checks. | Stars: 115k | Updated | Claude CodeCodex | Verified | Install |