Babysitter is a workflow engine for AI coding agents that enforces predefined steps, quality checks, human approvals, and decision records. It is used to coordinate complex, repeatable agent workflows across supported coding tools. The catalogue contains skills, agents, instructions, settings, a plugin, and an MCP integration for its workflow.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add a5c-ai/babysitter --skill fix-failing-pipelinesgit clone --depth 1 https://github.com/a5c-ai/babysitterWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/a5c-ai/babysitter/fix-failing-pipelines)<a href="https://agentmods.dev/skills/a5c-ai/babysitter/fix-failing-pipelines"><img src="https://agentmods.dev/badge/skills/a5c-ai/babysitter/fix-failing-pipelines/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/a5c-ai/babysitter/fix-failing-pipelines"><img src="https://agentmods.dev/badge/skills/a5c-ai/babysitter/fix-failing-pipelines.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.00970 |
| Opus 5 | $0.00030 | $0.00485 |
| Sonnet 5 | $0.00012 | $0.00194 |
| Haiku 4.5 | $0.00006 | $0.00097 |
Grade A, and why
fix-failing-pipelines scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 87 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fix Failing Pipelines
Check GitHub Actions workflows on the staging branch of https://github.com/a5c-ai/babysitter/actions, identify workflows whose most recent run is failing, and dispatch /babysitter:yolo to fix each one.
Workflow
Step 1: Fetch Most Recent Run Per Workflow
Use the gh CLI to list recent workflow runs on the staging branch:
gh run list --repo a5c-ai/babysitter --branch staging --limit 50 --json databaseId,workflowName,status,conclusion,createdAt,headBranch
Group the results by workflowName. For each workflow, keep only the most recent run (by createdAt). Discard workflows where the most recent run is still in_progress -- we only care about completed runs.
Step 2: Identify Failures
From the grouped results, select only workflows where the most recent completed run has conclusion: "failure". Skip workflows whose latest run succeeded, was cancelled, or is still running.
If no workflows have a failing most-recent run, report that all staging pipelines are green and stop.
Step 3: Get Failure Details
For each failing workflow run, fetch the failed job and step details:
gh run view <run_id> --repo a5c-ai/babysitter --json jobs --jq '.jobs[] | select(.conclusion == "failure") | {name, conclusion, steps: [.steps[] | select(.conclusion == "failure") | .name]}'
Then fetch the logs to understand the actual error:
gh run view <run_id> --repo a5c-ai/babysitter --log-failed 2>&1 | tail -100
Step 4: Present Failures
Display the list of failing workflows to the user with:
- Workflow name
- Run ID and link
- Failed job name(s) and failed step name(s)
- Brief summary of the error from the logs
Step 5: Fix via Babysitter
For each failing workflow, invoke the babysitter:yolo skill with a prompt that includes the failure context:
/babysitter:yolo fix the failing "<workflow_name>" pipeline on staging. The most recent run (<run_id>) failed in job "<job_name>" at step "<step_name>". Error details: <brief_error_summary>. Investigate the failure, fix the root cause, and push a fix to the staging branch. Do not create a new branch -- commit directly to staging.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 87 lines · 60 tokens per session scan A e09ee9d72f84
fix-failing-pipelines is a skill published in the GitHub repository a5c-ai/babysitter (1,788 stars, last pushed 6d ago), licensed MIT. It adds 60 tokens to every session and 970 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
principle-non-blocking-waits
Requires resumable waits for external state. Apply when waiting on CI, reviewers, deployments, or rate limits.
tutti-agent-workspace-app
Build or evolve a complex agent-enabled Tutti workspace app repository. Use for Tutti apps with web/server/shared monorepos, @tutti-os/agent-acp-kit local agent runtimes, kit-owned TUTTICLI agent/composer discovery, dynamic agent catalogs, run-scoped MCP tool gateways, app-owned package builders, web-first debugging…
pipe
Designing GitHub Actions workflows in depth: trigger strategy, security hardening, performance optimization, PR automation, and Reusable Workflow design.
goga-review-plan
Verify execution plan completeness and correctness.
goga-define-challenge
Challenge the complete product definition before the PRD is generated.
goga-define-goals
Define the product outcomes that should be achieved by solving the identified problem for the relevant users.