SkillAgentSearch skills...

greploop

Address Greptile feedback on a PR, MR, or shelved changelist through bounded review and fix cycles

Install / Use

npx skills add michaelshimeles/skills --skill greploop

Installs into whichever agent you are using.

About this skill
📄

SKILL.md

Installable skill definition

Quality Score

90/100

Category

Automation

Supported Platforms

Universal

Our assessment of greploop

greploop scores 90/100 on our quality scale, 1261st of 2,790 Automation skills we index (top 46%).

Its SKILL.md is 13 KB long, well organised into 20 sections with 31 code examples: a thorough specification that gives an agent plenty to work with.

With 1,076 GitHub stars, it is one of the more widely adopted skills in the catalogue.

Substance
30/30
Structure
20/20
Description
12/15
Adoption
13/20
Freshness
15/15

Maintenance, license and trust

  • The repository was last updated about 2 months ago, so greploop is actively maintained.
  • Our last check on 2026-10-30 found the source still online.
  • No license is declared. By default that means all rights are reserved: you can read it, but reusing or redistributing it is not clearly permitted. Ask the author before building on it commercially.
  • Its trust signals score 88/100, with 1 caution from licensing, adoption, age or documentation. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.

Safety scan

No issues found

Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.

AI review by kimi-k2.7-code on 2026-10-07. Automated pattern scan on 2026-10-07. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.

greploop compared with similar skills

All 4 of these similar skills score higher than greploop; compare them before choosing.

SkillScoreStarsUpdatedFormat
greploop (this skill)by michaelshimeles901.1k2mo agoSKILL.md
Agent-Reachby Panniantong100112.3k8d agoCLAUDE.md
rufloby ruvnet10081.9ktodayCLAUDE.md
headroomby headroomlabs-ai10079.8ktodayCLAUDE.md
Claude-Code-Game-Studiosby Donchitos10039.7k7d agoCLAUDE.md

Frequently asked questions

How do I install greploop?
Run npx skills add michaelshimeles/skills --skill greploop. The install tabs above show the steps for each supported agent.
Which AI agents does greploop work with?
It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
Is greploop safe to use?
Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It declares no license and scores 88/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
Is greploop still maintained?
The repository was last updated about 2 months ago, so greploop is actively maintained.

name: greploop description: > Iteratively improves a PR (GitHub), MR (GitLab), or shelved changelist (Perforce) until Greptile gives it a 5/5 confidence score with zero unresolved comments. Triggers Greptile review, fixes all actionable comments, pushes/re-shelves, re-triggers review, and repeats. Use when the user wants to fully optimize a PR/MR/CL against Greptile's code review standards. license: MIT compatibility: Requires git, gh (GitHub CLI) or glab (GitLab CLI) authenticated, and Greptile installed on the repo. For Perforce, requires p4 CLI authenticated. metadata: author: greptileai version: "1.3" allowed-tools: Bash(gh:) Bash(glab:) Bash(git:) Bash(p4:)

Greploop

Iteratively fix a PR/MR/CL until Greptile gives a perfect review: 5/5 confidence, zero unresolved comments.

Inputs

  • PR/MR/CL number (optional): If not provided, detect the PR/MR for the current branch, or the default pending changelist for p4.
  • --max-iterations N (optional, default 10): cap on review-fix-push cycles before the loop stops and reports the current state. Raise it for large PRs where Greptile keeps surfacing new findings; lower it to bound cost.

Instructions

0. Detect platform

First check for Perforce, then fall back to git remote detection:

# Check for Perforce environment
if p4 info >/dev/null 2>&1; then
  VCS="perforce"
else
  REMOTE_URL=$(git remote get-url origin)
  if echo "$REMOTE_URL" | grep -qi "gitlab"; then
    VCS="gitlab"
  else
    VCS="github"
  fi
fi

For self-hosted GitLab instances whose hostname doesn't contain "gitlab", the user can override by passing --vcs gitlab as an input. For Perforce, pass --vcs perforce.

1. Identify the PR/MR/CL

GitHub:

gh pr view --json number,headRefName -q '{number: .number, branch: .headRefName}'

GitLab:

glab mr view --output json | jq '{iid: .iid, branch: .source_branch}'

Switch to the PR/MR branch if not already on it.

Perforce:

# List pending changelists for current user/client
p4 changes -s pending -u $P4USER -c $P4CLIENT

# Describe a specific CL
p4 describe -s <CL_NUMBER>

Ensure the correct workspace (p4 client) is set before proceeding.

Key field differences:

  • GitHub: number, headRefName, headRefOid
  • GitLab: iid, source_branch, sha
  • Perforce: changelist number, P4CLIENT, shelved files

2. Loop

Repeat the following cycle at most --max-iterations times (default 10) to avoid runaway loops.

A. Trigger Greptile review

Push/shelve the latest changes (if any):

GitHub/GitLab:

git push

Perforce:

# Re-shelve to update the shelved files for review
p4 shelve -f -c <CL_NUMBER>

Wait for checks to start after push/shelve:

sleep 5

GitHub — check if Greptile is already running before posting a new trigger comment:

GREPTILE_STATE=$(gh pr checks <PR_NUMBER> --json name,state | jq -r '.[] | select(.name | test("greptile"; "i")) | .state')

If Greptile is not already running (PENDING or IN_PROGRESS), request a fresh review:

if [ "$GREPTILE_STATE" != "PENDING" ] && [ "$GREPTILE_STATE" != "IN_PROGRESS" ]; then
  gh pr comment <PR_NUMBER> --body "@greptile review"
fi

Then poll for the Greptile check run to complete:

HEAD_SHA=$(gh pr view <PR_NUMBER> --json headRefOid -q .headRefOid)
ATTEMPTS=0
MAX_ATTEMPTS=60
POLL_INTERVAL_SECONDS=10

while true; do
  ATTEMPTS=$((ATTEMPTS + 1))
  if [ "$ATTEMPTS" -gt "$MAX_ATTEMPTS" ]; then
    echo "Timed out waiting for the Greptile check run after approximately 10 minutes." >&2
    exit 1
  fi

  GREPTILE_CHECK=$(gh api "repos/{owner}/{repo}/commits/$HEAD_SHA/check-runs" \
    --jq '.check_runs[] | select(.name | test("greptile"; "i"))' 2>/dev/null)
  
  if [ -z "$GREPTILE_CHECK" ]; then
    echo "Waiting for Greptile check to appear..."
    sleep "$POLL_INTERVAL_SECONDS"
    continue
  fi
  
  STATUS=$(echo "$GREPTILE_CHECK" | jq -r '.status // "completed"')
  CONCLUSION=$(echo "$GREPTILE_CHECK" | jq -r '.conclusion // "pending"')
  
  if [ "$STATUS" = "completed" ]; then
    if [ "$CONCLUSION" = "success" ]; then
      echo "Greptile check passed!"
    else
      echo "Greptile check completed with: $CONCLUSION"
    fi
    break
  fi
  
  echo "Waiting for Greptile... (status: $STATUS)"
  sleep "$POLL_INTERVAL_SECONDS"
done

If polling times out, stop the greploop workflow and report the timeout. Do not continue with stale or missing review results.

GitLab — check if Greptile is already running before posting a trigger comment:

PIPELINES=$(glab api "projects/:fullpath/merge_requests/<MR_IID>/pipelines")
GREPTILE_RUNNING=$(echo "$PIPELINES" | jq '[.[] | select(.status == "running" or .status == "pending")] | length')

If no pipeline is running, post a trigger comment:

if [ "$GREPTILE_RUNNING" = "0" ]; then
  glab mr note <MR_IID> --message "@greptile review"
fi

Perforce — Perforce does not have native check runs. If Greptile is integrated via a webhook triggered on p4 shelve, wait for it to process. Check your Greptile installation's webhook endpoint or dashboard for the review status. Poll by re-fetching the Greptile review comment on the CL until a score appears.

Then poll for the Greptile pipeline job to complete (see GitLab API reference):

HEAD_SHA=$(glab mr view <MR_IID> --output json | jq -r '.sha')
ATTEMPTS=0
MAX_ATTEMPTS=60
POLL_INTERVAL_SECONDS=10

while true; do
  ATTEMPTS=$((ATTEMPTS + 1))
  if [ "$ATTEMPTS" -gt "$MAX_ATTEMPTS" ]; then
    echo "Timed out waiting for the Greptile pipeline job after approximately 10 minutes." >&2
    exit 1
  fi

  PIPELINES=$(glab api "projects/:fullpath/merge_requests/<MR_IID>/pipelines")
  # Find the most recent pipeline for this SHA
  PIPELINE_ID=$(echo "$PIPELINES" | jq -r --arg sha "$HEAD_SHA" \
    '[.[] | select(.sha == $sha)] | sort_by(.id) | last | .id // empty')

  if [ -z "$PIPELINE_ID" ]; then
    echo "Waiting for Greptile pipeline to appear..."
    sleep "$POLL_INTERVAL_SECONDS"
    continue
  fi

  JOBS=$(glab api "projects/:fullpath/pipelines/$PIPELINE_ID/jobs")
  GREPTILE_JOB=$(echo "$JOBS" | jq '.[] | select(.name | test("greptile"; "i"))')

  if [ -z "$GREPTILE_JOB" ]; then
    echo "Waiting for Greptile job to appear..."
    sleep "$POLL_INTERVAL_SECONDS"
    continue
  fi

  JOB_STATUS=$(echo "$GREPTILE_JOB" | jq -r '.status')

  if [ "$JOB_STATUS" = "success" ] || [ "$JOB_STATUS" = "failed" ] || [ "$JOB_STATUS" = "canceled" ]; then
    echo "Greptile job completed with: $JOB_STATUS"
    break
  fi

  echo "Waiting for Greptile... (status: $JOB_STATUS)"
  sleep "$POLL_INTERVAL_SECONDS"
done

If polling times out, stop the greploop workflow and report the timeout. Do not continue with stale or missing review results.

B. Fetch Greptile review results

Greptile may surface its score in several places — check all of the relevant sources:

GitHub:

1. PR description (body):

gh pr view <PR_NUMBER> --json body -q '.body'

2. General PR comments (issue comments):

gh api --paginate "repos/{owner}/{repo}/issues/<PR_NUMBER>/comments?per_page=100"

Filter for Greptile-authored comments and use the body from the most recently updated comment (updated_at), not the most recently created comment. Greptile may edit the same general PR comment on each review cycle; parse the current body, including the "Prompt to fix all with AI" section, before deciding there are no remaining issues.

3. PR reviews:

gh api repos/{owner}/{repo}/pulls/<PR_NUMBER>/reviews

Look for the most recent entry from greptile-apps[bot] or greptile-apps-staging[bot].

GitLab:

1. MR description (body):

glab mr view <MR_IID> --output json | jq -r '.description'

2. MR notes (comments):

glab api "projects/:fullpath/merge_requests/<MR_IID>/notes"

Filter for notes from the Greptile bot user (check the author.username field — the exact username may vary per installation; verify on first run).

Perforce:

1. CL description:

p4 describe -s <CL_NUMBER>

Check the description field for a Greptile-appended score block.

2. CL comments / review notes: If your installation uses a review tool such as Helix Swarm, fetch comments via its API.

Example (Swarm API): GET /api/v11/comments?topic=reviews/<REVIEW_ID>

Response fields of interest typically include:

  • user (author username)
  • body (comment text)
  • flags/state indicating whether the comment is resolved

Filter to comments authored by the Greptile bot:

  • Prefer exact username match if known
  • Otherwise, use a heuristic where the author name contains "greptile" (case-insensitive)

For all platforms, parse the text for:

  • Confidence score: a pattern like 3/5 or 5/5 (or Confidence: 3/5).
  • Comment count: Number of inline review comments noted in the summary.

Use whichever source has the most recently updated score. For GitHub, prefer updated_at from issue comments when comparing an edited Greptile summary against older review entries.

Also fetch all unresolved inline comments:

GitHub:

gh api repos/{owner}/{repo}/pulls/<PR_NUMBER>/comments

Also carry forward actionable items from the latest Greptile general PR comment, especially the "Prompt to fix all with AI" section, even if the inline comment endpoint returns zero unresolved comments.

GitLab:

glab api "projects/:fullpath/merge_requests/<MR_IID>/discussions"

Filter to DiffNote type discussions (notes[0].type == "DiffNote") from Greptile that are on the latest commit and not yet resolved ("resolved": false).

Perforce: If using Swarm:

Fetch inline diff comments for the review associated with the CL

GET /api/v11/comments?topic=reviews/<REVIEW_ID>

Filter to comments from the Greptile bot user that have not been marked as resolved/addressed.

C. Check exit conditions

Stop the loop if any of these are true:

  • Confidence score is 5/5 AND there are zero unresolved comments
  • --max-iterations reached (report current state)

D. Fix actionable comments

For each unresolved Greptile comment:

  1. Read the file and understand the comment in context.
  2. Determine if it's actionable (code change needed) or informational.
  3. If actionable, make the fix.
  4. If informational or a false positive, note it but still resolve the thread.

E. Resolve threads

GitHub — fetch unresolved review threads and resolve all that have been addressed (see GraphQL reference):

gh api graphql -f query='
query($cursor: String) {
  repository(owner: "OWNER", name: "REPO") {
    pullRequest(number: PR_NUMBER) {
      reviewThreads(first: 100, after: $cursor) {
        pageInfo { hasNextPage endCursor }
        nodes {
          id
          isResolved
          comments(first: 1) {
            nodes { body path author { login } }
          }
        }
      }
    }
  }
}'

Resolve addressed threads:

gh api graphql -f query='
mutation {
  t1: resolveReviewThread(input: {threadId: "ID1"}) { thread { isResolved } }
  t2: resolveReviewThread(input: {threadId: "ID2"}) { thread { isResolved } }
}'

GitLab — fetch unresolved discussions and resolve each one (see GitLab API reference):

glab api "projects/:fullpath/merge_requests/<MR_IID>/discussions?per_page=100"

Filter for "resolved": false discussions. Then resolve each by its id:

glab api --method PUT \
  "projects/:fullpath/merge_requests/<MR_IID>/discussions/<DISCUSSION_ID>" \
  --field resolved=true

Repeat for each unresolved discussion ID. (GitLab has no batch resolution — loop through each one.)

F. Commit and push / re-shelve

GitHub/GitLab:

git add -A
git commit -m "address grepti

Truncated for display — read the full file on GitHub.

Related Skills

View on GitHub
GitHub Stars1.1k
CategoryAutomation
Updated2mo ago
Forks181

Languages

Python

Trust signals

88/100

From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.

1 medium