Skip to main content
ClaudeWave
Install in Claude Code
Copy
git clone --depth 1 https://github.com/jeremylongshore/tons-of-skills-marketplace /tmp/sticky && cp -r /tmp/sticky/plugins/ai-agency/hyperflow/skills/sticky ~/.claude/skills/sticky
Then start a new Claude Code session; the skill loads automatically.

SKILL.md

# Sticky

Set per-project auto-routing mode. Three states:

| State | Default? | Behavior |
|---|---|---|
| `auto` | yes (when `.sticky` absent) | **Intent-detection routing** — messages containing chain-starter verbs (`audit`, `debug`, `fix`, `brainstorm`, `scope`, `deploy`, `review`, …) auto-route. Pure conversation passes through. |
| `on` | — | **Full sticky** — every task-shaped message routes, even without explicit intent verbs |
| `off` | — | **All auto-routing disabled** — only explicit `/hyperflow:*` slash commands trigger chains |

Intent-detection is the floor — the user gets it without any opt-in (the orchestrator scans every user message for chain-starter verbs and routes when matched). Sticky `on` raises the ceiling; sticky `off` lowers the floor.

Full doctrine: [DOCTRINE.md](../hyperflow/DOCTRINE.md) Layer 1 auto-routing clause (intent verb taxonomy + routing contract + bypass patterns).

## Subcommands

| Subcommand | Description |
|---|---|
| `on` | Set state: on — full sticky routing on every task-shaped message |
| `auto` | Set state: auto — intent-verb routing only (the default) |
| `off` | Set state: off — disable ALL auto-routing including intent detection |
| `status` | Show current state (on / auto / off / when toggled) |

Default subcommand when none provided: `status`.

## State persistence

Sticky state is stored at `.hyperflow/.sticky` (project-scoped, gitignored). File format:

```
state: on
since: 2026-05-17T14:30:00Z
trigger: user-mention   # or: explicit-toggle | session-default
```

The session-start hook reads this file and prints a one-line advisory when sticky is on (`Sticky mode: ON since 2026-05-17 14:30 — task-shaped messages auto-route through hyperflow`). Sticky persists across sessions until explicitly toggled off.

## Subcommand Details

### `on`

Write `.hyperflow/.sticky` with `state: on` + ISO-8601 timestamp + `trigger: explicit-toggle`. Print:

```
Sticky mode: ON (full routing)
Every task-shaped message now routes through hyperflow, even without intent verbs.
Disable with /hyperflow:sticky off · or relax to verb-only routing with /hyperflow:sticky auto.
```

### `auto`

Write `.hyperflow/.sticky` with `state: auto` + timestamp. This is the default state when no file exists; explicitly setting it is useful after `off` to re-enable intent-detection without going to full sticky. Print:

```
Sticky mode: AUTO (intent-detection routing, default)
Messages containing chain-starter verbs (audit, debug, fix, brainstorm, scope, deploy, review, …) auto-route.
Pure conversation passes through. Expand to full routing with /hyperflow:sticky on.
```

### `off`

Replace `.hyperflow/.sticky` contents with `state: off` + timestamp. (Keep the file rather than delete so the session-start hook can show recent history.) Print:

```
Sticky mode: OFF
All auto-routing disabled — even intent verbs (audit, debug, fix, brainstorm, …) will no longer route.
Use explicit /hyperflow:* invocations. Re-enable with /hyperflow:sticky auto or /hyperflow:sticky on.
```

### `status`

Read `.hyperflow/.sticky`. Print one line:

```
Sticky mode: ON since 2026-05-17 14:30 (trigger: user-mention)
```

or:

```
Sticky mode: AUTO since 2026-05-17 14:30 (trigger: default · intent-detection routing)
```

or:

```
Sticky mode: OFF (last changed: 2026-05-16 09:12)
```

or, if file absent:

```
Sticky mode: AUTO (default · file not yet written · intent-detection routing active)
```

## Behavioural contract

When sticky is ON, the orchestrator MUST follow this routing on every new user message:

1. **Chat-shaped messages** (questions about prior output, "yes" / "no" answers to a pending gate, acknowledgments like "ok"/"thanks", short clarifications) — pass through normally, no chain routing.
2. **Task-shaped messages** (any verb-led request for new work: "add X", "fix Y", "refactor Z", "build", "implement", "create", "design", "scope out", "decompose", "ship") — auto-route:
   - **New work** (whether the user asks *what* / *should we* or describes *how* and names concrete files / functions) → invoke `/hyperflow:plan` with the user's message as `ARGUMENTS`. Plan internally bounces past the design phase straight to decomposition when the request is already clear.
   - **Existing task file referenced** (e.g. "resume the auth task") → invoke `/hyperflow:dispatch` with the matching slug.
3. **Bug reports** ("X is broken", "Y test fails", "Z throws…") → invoke `/hyperflow:trace`.
4. **Review requests** ("review this", "audit the diff", "any issues?") → invoke `/hyperflow:audit`.
5. **Ship intent** ("ship it", "push", "release", "deploy") → invoke `/hyperflow:deploy`.

The routing decision is made silently — print one short line (`Routing to /hyperflow:plan (sticky mode) …`) and invoke. Don't ask the user to confirm the routing (that would be an invented gate per DOCTRINE rule 8). The Step 0 chain-mode question still fires inside the routed skill.

**Override:** if the user message starts with `/` (any slash command) OR contains "without hyperflow" / "skip hyperflow" / "don't route" → bypass routing for that message; respond directly.

## Activation triggers

Intent-detection routing (`state: auto`) is the **default** — active for every project without any user action. The orchestrator scans every user message for chain-starter verbs (per the DOCTRINE intent verb taxonomy) and routes when matched. No file write needed.

Upgrades and downgrades:

1. **Upgrade to full sticky (`on`):**
   - Explicit: user runs `/hyperflow:sticky on`.
   - Implicit: user mentions the word "hyperflow" in a non-slash-command message AND `.hyperflow/.sticky` does not exist OR is `auto`. Orchestrator writes `state: on · trigger: user-mention · since: <ISO-8601>` and prints `Sticky mode: ON (upgraded from auto, activated by mention). Disable with /hyperflow:sticky off.`
2. **Downgrade to intent-only (`auto`):** user runs `/hyperflow:sticky auto`.
3. **Disable entirely (`off`):** user runs `/hyperflow:sticky off`. Disabl
beads-wardenSubagent

Guard the beads execution record: enforce the write-flush-verify discipline that defeats the bd rapid-write race, audit epic dependency graphs for cycles and orphans, catch closures whose title overstates what shipped, flag open beads carrying no disposition or a disproven premise, and reconcile bd against its GitHub and Plane projections. Owns RECORD INTEGRITY; delegates graph analysis to bead-dependency-mapper and epic-closure drift to bead-epic-auditor rather than duplicating them. Use before closing an epic, after any batch of bd writes, when a bead premise looks stale, or when auditing whether the record matches reality. Trigger with "audit beads", "check the bead DAG", "did that close actually land", "bead hygiene".

claim-verifierSubagent

Verify every factual assertion in a diff, PR body, commit message, bead note, or governing doc against the actual repository, and fail anything that cannot be substantiated by a command. Use before merging any PR that makes claims about counts, coverage, consumers, enforcement, provenance, or certification, and when auditing standing docs for rot. Trigger with "verify claims", "check this PR body", "is this claim true", "claim audit".

omarchy-plugin-architectSubagent

Design and build Omarchy (Quickshell/QML) bar-widget, panel, and service plugins that actually work on a stock install. Knows the hard runtime constraint (no node on the graphical session PATH), the first-party contracts (BarWidget, Panel, KeyboardPanel, PanelKeyCatcher, Service), the curl-from-QML data pattern, FileView persistence, and the marketplace submission bar. Use when starting a new Omarchy plugin, porting a plugin off an external runtime, wiring a service to a bar widget, or deciding how a widget should fetch and persist. Trigger with "build an omarchy plugin", "omarchy widget", "quickshell plugin", "port this plugin to QML".

omarchy-submission-auditorSubagent

Audit an Omarchy plugin before it reaches the marketplace: prove it installs and runs on a stock box (no node/python on the session PATH), run the omarchy-submit gate lane, validate on the rig with omarchy-plugin-validate and qmllint, and check the QML security invariants and first-party idiom contracts. Read-only: it reports and blocks, it does not rewrite the plugin. Use before submitting an entry, after any data-layer change, or when a plugin works on the dev box and you need to know whether it works for a real user. Trigger with "audit this omarchy plugin", "is this plugin submission ready", "will this plugin work when installed".

skill-auditorSubagent

Audit and fix Claude Code SKILL.md files against enterprise compliance standards: frontmatter completeness, required body sections, and style. Use when validating or repairing skills in a plugin directory. Trigger with "audit skill", "fix skill compliance".

getting-startedSkill

Learn how SKILL.md files work in Claude Code plugins, then build a production-quality agent skill from scratch. Covers frontmatter schema, body structure, testing, and iteration.

guidesSkill

Step-by-step guide to writing a SKILL.md file for Claude Code. Learn how to plan, structure, and test auto-activating skills with proper frontmatter, allowed-tools, dynamic context injection, and supporting files.

agency-osSkill

|