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 self-optimizationgit 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/self-optimization)<a href="https://agentmods.dev/skills/a5c-ai/babysitter/self-optimization"><img src="https://agentmods.dev/badge/skills/a5c-ai/babysitter/self-optimization/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/self-optimization"><img src="https://agentmods.dev/badge/skills/a5c-ai/babysitter/self-optimization.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 | $0.00032 | $0.00361 |
| Opus 5 | $0.00016 | $0.00180 |
| Sonnet 5 | $0.00006 | $0.00072 |
| Haiku 4.5 | $0.00003 | $0.00036 |
Grade A, and why
self-optimization 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 9d 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.
What it actually says
- Improving routing and agent selection over time
- Adapting to new project patterns without forgetting old ones
- Building cross-session intelligence
SONA Cycle
- Extract Patterns - Mine execution data for recurring patterns
- RETRIEVE - Search ReasoningBank for matching trajectories
- JUDGE - Evaluate trajectory applicability in current context
- DISTILL - Compress and store new entries
- Adapt - Update weights with EWC++ regularization
Anti-Forgetting (EWC++)
- Elastic Weight Consolidation prevents overwriting previously learned patterns
- Fisher information matrix tracks parameter importance
- Configurable regularization penalty for new adaptations
RL Algorithms
Q-Learning, SARSA, PPO, DQN, A2C, TD3, SAC, DDPG, Rainbow
Agents Used
agents/optimizer/- Performance tuningagents/adaptive-queen/- Real-time adaptation
Tool Use
Invoke via babysitter process: methodologies/ruflo/ruflo-intelligence
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 9d ago First seen · 43 lines · 32 tokens per session scan A 7fdf372ed802
self-optimization is a skill published in the GitHub repository a5c-ai/babysitter (1,788 stars, last pushed 7d ago), licensed MIT. It adds 32 tokens to every session and 361 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
reflect
Mines a session for durable learnings. Trigger on "reflect on this session" or "/reflect" only; never infer intent from session end or friction.
list-learned-actions
Explicit Codex workflow: List persisted reusable actions, UI skeletons, and legacy feedback memories before composing device primitives.
agent-memory-coordinator
Agent skill for memory-coordinator - invoke with $agent-memory-coordinator.
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…
distill-session-knowledge
Offline-mine this project's pi session JSONL logs into reusable, verified knowledge: extracts faults, decisions, corrections, procedures and docs, promotes only recurring patterns, and routes artifacts into skillmanage, memory and docs. Use on "mine my sessions", "distill session knowledge", "extract lessons from…
nest
Designing LLM-optimized folder structures: audits and restructures directories for context efficiency, progressive disclosure, and prompt cache performance. Not for general repo structure (Grove).