
- Installs
- 783
- Last 24h
- —
- Trending rank
- —
- Security
- Not audited
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/boshu2/agentops --skill ratchet- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads ratchet automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
SKILL.md
The instructions your agent reads when the skill is triggered.
Ratchet Skill
Track progress through the RPI workflow with permanent gates.
Note: /ratchet tracks and locks progress. It does not “run the loop” by itself—pair it with /crank (epic loop) or /swarm (Ralph loop) to actually execute work.
The Brownian Ratchet
Progress = Chaos × Filter → Ratchet
| Phase | What Happens |
|---|---|
| Chaos | Multiple attempts (exploration, implementation) |
| Filter | Validation gates (tests, /vibe, review) |
| Ratchet | Lock progress permanently (merged, closed, stored) |
Key insight: Progress is permanent. You can't un-ratchet.
Execution Steps
Given /ratchet [command]:
status - Check Current State
ao ratchet status 2>/dev/null
Or check the chain manually:
cat .agents/ao/chain.jsonl 2>/dev/null | tail -10
check [step] - Verify Gate
ao ratchet check <step> 2>/dev/null
Steps: research, plan, implement, vibe, post-mortem
record [step] - Record Completion
ao ratchet record <step> --output "<artifact-path>" 2>/dev/null
Or record manually by writing to chain:
echo '{"step":"<step>","status":"completed","output":"<path>","time":"<ISO-timestamp>"}' >> .agents/ao/chain.jsonl
skip [step] - Skip Intentionally
ao ratchet skip <step> --reason "<why>" 2>/dev/null
Workflow Steps
| Step | Gate | Output |
|---|---|---|
research | Research artifact exists | .agents/research/*.md |
plan | Plan artifact exists | .agents/plans/*.md |
implement | Code + tests pass | Source files |
vibe | /vibe passes | .agents/vibe/*.md |
post-mortem | Learnings extracted | .agents/learnings/*.md |
Chain Storage
Progress stored in .agents/ao/chain.jsonl:
{"step":"research","status":"completed","output":".agents/research/auth.md","time":"2026-01-25T10:00:00Z"}
{"step":"plan","status":"completed","output":".agents/plans/auth-plan.md","time":"2026-01-25T11:00:00Z"}
{"step":"implement","status":"in_progress","time":"2026-01-25T12:00:00Z"}
Key Rules
- Progress is permanent - can't un-ratchet
- Gates must pass - validate before proceeding
- Record everything - maintain the chain
- Skip explicitly - document why if skipping a step
Examples
Check RPI Progress
User says: /ratchet status
What happens:
- Agent calls
ao ratchet status 2>/dev/nullto check current state - CLI reads
.agents/ao/chain.jsonland parses progress - Agent reports which steps are completed, in-progress, or pending
- Agent shows output artifact paths for completed steps
- Agent identifies next gate to pass
Result: Single-screen view of RPI workflow progress, showing which gates passed and what's next.
Record Step Completion
Skill says: After /research completes
What happens:
- Agent calls
ao ratchet record research --output ".agents/research/auth.md" - CLI appends completion entry to
.agents/ao/chain.jsonl - Agent locks research step as permanently completed
- Agent proceeds to plan phase knowing research gate passed
Result: Progress permanently recorded, gate locked, workflow advances without backsliding.
Troubleshooting
| Problem | Cause | Solution |
|---|---|---|
| ao ratchet status fails | ao CLI not available or chain.jsonl missing | Manually check .agents/ao/chain.jsonl or create empty file |
| Step already completed error | Attempting to re-ratchet locked step | Use ao ratchet status to check state; skip if already done |
| chain.jsonl corrupted | Malformed JSON entries | Manually edit to fix JSON; validate each line with jq -c '.' <file> |
| Out-of-order steps | Implementing before planning | Follow RPI order strictly; use --skip only with explicit reason |
Reference Documents
- references/ratchet.feature — Executable spec: Chaos×Filter→Ratchet, record/check loop-step gates, monotonic chain (soc-qk4b.2)
FAQ
What does the ratchet skill do?
Record Brownian Ratchet gates.
How do I install ratchet?
Run npx skills add https://github.com/boshu2/agentops --skill ratchet in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/ratchet/ (personal) or .claude/skills/ratchet/ (project).
Which AI agents can use ratchet?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is ratchet safe to install?
No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.
Who maintains ratchet?
It is published by boshu2 in boshu2/agentops. It has 783 recorded installs.
More from boshu2/agentops
council
boshu2/agentops
Compare independent opinions from several models or contexts without inflating agreement. Use when: wanting a second opinion or debate, or summarizing several reviewers' results.
doc
boshu2/agentops
Write source-checked docs, READMEs, change descriptions and handoffs. Use when: documenting a change, leaving or resuming a handoff, or drafting a questionnaire for someone else.
research
boshu2/agentops
Answer one cited question on code, history, claims or patterns. Use when: asked how it works, why it changed, if a claim or done/shipped report holds, or a pattern warrants a rule.
plan
boshu2/agentops
Shape a request into one end-to-end slice with observable behavior; review write scope and reversible decisions. Use when: planning, breaking down or scoping a change.
swarm
boshu2/agentops
Dispatch explicit disjoint packets exactly once through a caller-selected executor. Triggers: "swarm", "dispatch disjoint packets", "parallel explicit tasks".
rpi
boshu2/agentops
Drive one accepted change through implementation and checks to done, with one fresh review only where a mistake is costly. Use when: asked to run or orchestrate work through RPI.
Popular other skills
hyperframes-cli
heygen-com/hyperframes
The HyperFrames CLI by job - scaffold, check, look at frames, edit the timeline, preview, render locally or in the cloud, publish, project history - plus reading its JSON and recovering from build or render failures.
triage
mattpocock/skills
Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs.
ask-matt
mattpocock/skills
Ask which skill or flow fits your situation. A router over the skills in this repo.
to-questionnaire
mattpocock/skills
Turn a decision you can't fully answer into a questionnaire for someone else to fill in.
loop-me
mattpocock/skills
Grill me about specs for the workflows I want to build, within this workspace.