What makes claudify notable?

2.1K downloads. Trust evidence available

How should claudify be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

claudify

Agentic AI lifecycle: create + improve + persist. create - convert functionality into Claude Code automation (agent/skill/rule/hook/command) [SKILL.md], background-polling - mandatory ScheduleWakeup/timeout polling discipline for 5min+ background dispatches [background-polling.md], improve - self-improving loop: retrospect + hook/skill review + pattern detect [improve.md], persist - knowledge persistence: documentation + memory save [persist.md]. Use when "agentify", "agentic", "automate this", "create an agent", "make a plugin", "make a skill", "self-improve", "claudify improve", "claudify persist", "background polling", "ScheduleWakeup".

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

2.1k

Updated

Oct 9, 2026

Version

0.10.0

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 2.1K downloads reported by the source. Last updated 10/9/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 9, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 9, 2026
Adoption signal: 2.1K downloadsadoption · observed Oct 9, 2026
Latest release: 0.10.0release · observed Sep 14, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17ay1v6v88r2m102pvvc44gz183qcrm:claudify

Install using `clawhub skill install s17ay1v6v88r2m102pvvc44gz183qcrm:claudify` in an isolated environment before connecting it to live workloads.
No published capability contract is available yet, so validate auth and request/response behavior manually.
Review the upstream CLAWHUB listing at https://clawhub.ai/drumrobot/claudify before using production credentials.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-drumrobot-claudify/snapshot"

Documentation

CLAWHUB

152,959 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: claudify
description: |
  Agentic AI lifecycle: create + improve + persist. create - convert functionality into Claude Code automation (agent/skill/rule/hook/command) [SKILL.md], background-polling - mandatory ScheduleWakeup/timeout polling discipline for 5min+ background dispatches [background-polling.md], improve - self-improving loop: retrospect + hook/skill review + pattern detect [improve.md], persist - knowledge persistence: documentation + memory save [persist.md]. Use when "agentify", "agentic", "automate this", "create an agent", "make a plugin", "make a skill", "self-improve", "claudify improve", "claudify persist", "background polling", "ScheduleWakeup".
metadata:
  author: es6kr
  version: "0.1.2"
allowed-tools:
  - Read
  - Write
  - Edit
  - Glob
  - Skill
  - Bash(mkdir:*)
---

# Claudify

Guide users to convert functionality into the appropriate Claude Code automation type (Agent, Skill, Rule, Command, or Hook).

## Topic Dispatch

**When this skill is invoked with a topic specifier (e.g., `Skill("claudify", "improve")` or `/claudify improve`), load and follow ONLY the matching topic file below — do NOT default to the creation-wizard body (Decision Matrix / Workflow) in this SKILL.md.** A bare invocation (no topic, `agentify`, or `create`) uses the creation workflow below.

| Topic arg | File | Purpose |
|-----------|------|---------|
| `create` / `agentify` / (none) | this SKILL.md (below) | Convert functionality into an automation (Agent/Skill/Rule/Command/Hook) |
| `improve` | [improve.md](./improve.md) | Self-improving loop: retrospect + hook/skill review + pattern detect |
| `persist` | [persist.md](./persist.md) | Knowledge persistence: documentation + memory save |
| `background-polling` | [background-polling.md](./background-polling.md) | ScheduleWakeup/timeout polling discipline for 5min+ background dispatches |

For `improve` / `persist` / `background-polling`, **Read that topic file and follow it** — the Decision Matrix and creation Workflow below apply only to the `create`/`agentify`/(none) creation path. For any other/unrecognized topic arg, fall back to the `create`/`agentify`/(none) creation workflow below rather than assuming a matching topic file exists.

## Decision Matrix

| Type | When to use | Implementation |
| ---- | ----------- | -------------- |
| **Agent** | High autonomy, multi-tool | `.claude/agents/name.md` |
| **Skill** | Domain expertise, logic | `.claude/skills/name/SKILL.md` |
| **Rule** | Constraints, styling | `.claude/rules/name.md` |
| **Slash Command** | User types `/cmd` | Simple prompt templates |
| **Hook** | Events (tool use, etc) | Automation on actions |

Full comparison: [automation-decision-guide.md](./resources/automation-decision-guide.md)

## Workflow

### Step 0: Output-size routing (HARD STOP — before Step 1)

**Before any large-output operation, decide: inline vs subagent vs script.** Parent context bloat from claudify itself reduces the budget available for the actual

_meta.json

{
  "ownerId": "kn74k8yfvftx6f062qa8fzyd8h8373jd",
  "slug": "claudify",
  "version": "0.10.0",
  "publishedAt": 1789395318834
}

background-polling.md

# Background Polling

Mandatory active polling for long-running background work (HARD STOP).

**Immediately after dispatching a background task expected to take 5+ minutes, register a `ScheduleWakeup` or actively poll via `Monitor`.** Never sit idle until the user asks "is it done yet?".

## Notification / polling matrix by dispatch type

| Dispatch method | Automatic notification | Active polling required? |
|-----------------|------------------------|--------------------------|
| `Agent(run_in_background: true)` | ✅ task-notification on completion (but never arrives if the agent hangs) | ⚠️ Register ScheduleWakeup when 5+ min expected |
| `Bash(run_in_background: true)` (command includes a timeout) | ✅ task-notification on completion/failure | ⚠️ Register ScheduleWakeup when 5+ min expected |
| **`Bash(run_in_background: true)` (no timeout — HARD STOP)** | ❌ Never notified on hang (the tool `timeout` parameter does NOT apply to background) | **✅ `timeout N <cmd>` prefix in the command itself is mandatory. Without it, background dispatch is forbidden** |
| **Remote-host nohup detach (e.g. `ssh remote-host 'nohup ... &'`)** | ❌ No automatic notification | **✅ Required** — register `ScheduleWakeup` |
| **External-system async work (CI run, cloud deploy, etc.)** | Partial (`gh run watch` etc.) | **✅ Required** if no native watch |
| **User-triggered CI/deploy (user pushes a tag, runs manual workflow_dispatch, or fires an external trigger)** | ❌ No automatic notification | **✅ Required** — poll primary sources (`gh run list` / `gh release list`) immediately before composing the next response |
| **Assistant-triggered PR creation (`gh pr create` fires CI on the PR)** | ❌ No automatic notification | **✅ Required** — `gh pr checks <N> --watch` (or non-blocking poll if other work is drivable), then `gh pr ready <N>` on green. This is part of the same PR-creation task, not a deferrable follow-up (`github-flow/pr.md` Step 7.5) |

## `TaskOutput`'s own `block`/`timeout` — same short-cycle discipline applies (HARD STOP)

**`TaskOutput(block=true, timeout=<large>)` is a synchronous in-turn wait, not a background dispatch — it burns the exact same wall-clock/cache-window cost as an idle loop, and it is invisible to the Bash-only ceiling check (`bash-guard.py` `TIME_BOUND_CEILING`) and to the Stop-hook wait-phrase detector (which only fires when a turn *ends* on a wait). A multi-minute `block=true` call never ends the turn, so neither guard sees it.**

- `block=true` is for a **short confirmation check** (≤30s) right after dispatch — not a way to "wait out" an expected multi-minute task inside one call.
- If the first short check returns `not_ready`/`running`, do **not** re-issue another near-ceiling blocking call. Either: (a) end the turn and register `ScheduleWakeup` per the delay guide above, or (b) if other pending work exists this turn, drive it now and re-check with `block=false` later.
- `block=false` (non-blocking poll) is always safe regardless of dur

CHANGELOG.md

# Changelog

## [0.10.0](https://github.com/es6kr/skills/compare/claudify-v0.9.0...claudify-v0.10.0) (2026-09-13)


### Features

* commit-tag package-count guard + hook-kit relocations (promotes develop batch to main) ([a9a5282](https://github.com/es6kr/skills/commit/a9a5282489301412405a2f1024a396ba17da1e48))


### Bug Fixes

* **claudify:** stop naming the private companion repo in improve.md ([b24038c](https://github.com/es6kr/skills/commit/b24038c0c10cc156ee8947d90b070102181b7082))
* **hook-kit:** relocate hook-registry.yaml and its tooling to es6kr/claude-plugins ([6407d6c](https://github.com/es6kr/skills/commit/6407d6cc3f7c26ded22b1812e322ad37131b74f4))

## [0.9.0](https://github.com/es6kr/skills/compare/claudify-v0.8.0...claudify-v0.9.0) (2026-09-05)


### Features

* promote next-feat (fa lifecycle ownership) ([90c90bb](https://github.com/es6kr/skills/commit/90c90bb4796087ff6eaa5491b3c4524f7ed829a7))


### Bug Fixes

* **claudify:** dispatch FA prune through Skill("fa", "fa-prune") ([82468bf](https://github.com/es6kr/skills/commit/82468bf455eed29bbb260d65a15ae24727779099))
* **claudify:** make improve B.1 hook review registry-first ([2a24e73](https://github.com/es6kr/skills/commit/2a24e73bfb92ce7704b8f210c6944f0dde4e7cda))

## [0.8.0](https://github.com/es6kr/skills/compare/claudify-v0.7.1...claudify-v0.8.0) (2026-09-01)


### Features

* **claudify:** require retrospect findings to reach the tracker ([d1703b5](https://github.com/es6kr/skills/commit/d1703b55090b57193c659ec4e403eb4307a893e4))
* staging branch next-feat sync into main ([e582ea2](https://github.com/es6kr/skills/commit/e582ea248beefa70886716741b8965d91cfb7414))

## [0.7.1](https://github.com/es6kr/skills/compare/claudify-v0.7.0...claudify-v0.7.1) (2026-08-26)


### Bug Fixes

* accumulate 16 patch-level bug fixes and guard enhancements across skills ([d214e5d](https://github.com/es6kr/skills/commit/d214e5dcc7fac1bc07baf3b6cec62999aea732f0))
* **core:** align workflow steps, next suggestion patterns, and browser topics ([c68d489](https://github.com/es6kr/skills/commit/c68d489d01a79862b8933b4a0542168cf676cd3a))
* promote next-fix batch (consolidate fabrication guard, session rewind, config-driven PR base) ([7ca0ccb](https://github.com/es6kr/skills/commit/7ca0ccbf13cefafedc33a16a7361756c95f8b8f6))

## [0.7.0](https://github.com/es6kr/skills/compare/claudify-v0.6.1...claudify-v0.7.0) (2026-08-20)


### Features

* promote next-feat staging (lifecycle guards, triage automation, and workflow safety procedures) ([77d58ac](https://github.com/es6kr/skills/commit/77d58ac3a771a4897043c9eea8b149ea1e8ba2ff))

## [0.6.1](https://github.com/es6kr/skills/compare/claudify-v0.6.0...claudify-v0.6.1) (2026-08-17)


### Bug Fixes

* promote next-fix staging (30 fixes across 14 skills) ([ee467c0](https://github.com/es6kr/skills/commit/ee467c045d779d7b80d30f160763ec3534a9742b))
* **wip:** cross-ref PR-URL and TaskCreate subject repo-qualifier rules ([#186](https://github.com/es6kr/skills/issues/186

improve.md

# Improve (Self-Improving Loop)

Analyze session episodic data (mistakes, hook/skill behavior, repeated patterns) and improve the system.

## When to Use

- `/claudify improve` — direct invocation
- `/cleanup run` Step 2 — automatic as part of session cleanup

## Workflow

### A. Retrospect (Mistake Analysis)

Scan conversation for mistake signals and record to feedback memory + failed-attempts.md.

The record procedure itself is owned by the `fa` skill — invoke `Skill("fa")` (its
[retrospect.md](../fa/retrospect.md) topic) for the actual write; this section only defines
the improve-loop framing around it.

**Signals**:

| Signal | Example |
|--------|---------|
| User correction | "no, not that", "you didn't even verify?" |
| Wrong judgment | Model declared absent, guessed separate issue |
| Ignored artifacts | Existing plan/research files not checked |
| Repeated mistake | Same correction 2+ times |
| Rule violation | Performed prohibited action from rules |

**Per mistake**:
1. Analyze root cause (Why 1-3 minimum)
2. Draft feedback memory entry (rule + Why + How to apply)
3. **If the finding implies future work, register it as a `- [ ]` item in the workspace tracker (`fix_plan.md` / `checklist.md`) — HARD STOP.** See below.
4. Collect for Phase 2 AskUserQuestion

**Findings that imply future work must reach the tracker (HARD STOP)**

The retrospect log records **what went wrong**; the tracker records **what will be done**. They are different media and one does not stand in for the other. A finding whose remediation is not fully executed in this session — a hook the escalation matrix now mandates, a trigger to register, a rule to strengthen, a defect to verify elsewhere — leaves no actionable trace when it lives only in the retrospect entry: nothing surfaces it during the next session's backlog read, so the escalation the entry itself declares never gets executed.

This mirrors the obligation C (Pattern Detect) already carries for its candidates. The asymmetry — candidates registered, retrospect remediations not — is the gap this rule closes.

| # | Don't | Do |
|---|-------|-----|
| 1 | Write the retrospect entry, then treat the retrospect as done | Register every finding's outstanding remediation as a `- [ ]` tracker item, cross-referenced by the entry's class name |
| 2 | Route the "should we do this remediation?" decision to a user ask instead of the tracker | Registering it is not asking to do it — it makes the work visible. Whether to run it now is a separate decision |
| 3 | Skip registration because the retrospect entry already spells out the fix in its Why 5 | A retrospect entry is not read during backlog triage. If it is not in the tracker, it is not scheduled |
| 4 | Register only findings from Pattern Detect because that is where the HARD STOP is written | Retrospect findings carry the same obligation — this section is that missing half |

**Skip condition for registration only**: the finding's remediation was fully executed in this

Github ReposUpdated 8h agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.