erne-worker

Engineering Community

Repo stars 36
License MIT
Author repo everything-react-native-expo

Interpretation is structured for decision-making; original keeps the upstream SKILL.md unchanged.

Decide Fit First

Core job: ERNE — Autonomous ticket execution — polls a provider, picks up ready tickets, and runs the full ERNE pipeline (validate, plan…
Best fit: Use it when the task has reusable inputs, steps, and validation criteria rather than a one-off answer.
Avoid forcing it: If the source lacks commands, platform support, or external-service evidence, keep those fields unknown instead of guessing.

Design Intent

Structure: The skill is organized around “Usage”, “Options”, “Pipeline Steps”, “Supported Providers”, showing how the author expects the agent to judge fit, collect context, and produce verifiable output.
Trigger evidence: Prioritize the author’s wording around when to use it, what context to collect, and what output shape to produce.
Evidence boundary: Author text states facts, repository files prove commands and paths, and Fluxly only adds fit, limits, and usage judgment.

How To Use It

Inputs: Provide target material, scope, expected result, forbidden changes, and validation method.
Invocation: Name erne-worker directly; if the source includes slash commands, start with the command and then add task context.
Validation: Start small and check whether the result follows “Usage / Options / Pipeline Steps” before expanding.

Boundaries And Review

Dependencies: Prepare Vendor-specific API keys before running a full task.
Permissions: Declared permissions include read / write / shell-exec; ask the agent to state file, command, and rollback boundaries before acting.
Quality bar: A useful result names the deliverable, evidence, and next action. Generic prose means the task needs tighter context.

/erne-worker — Autonomous Ticket Execution

Usage

erne worker --config <path-to-worker.json>

Options

Flag	Description
`--config <path>`	Path to worker configuration JSON (required)
`--dry-run`	Fetch tickets and print them without executing
`--once`	Process one ticket and exit

Pipeline Steps

Poll — Fetch ready tickets from the configured provider
Validate — Check ticket has enough detail (title, description, acceptance criteria)
Confidence Score — Estimate likelihood of autonomous success (0-100)
Context Resolve — Load project stack, audit data, and relevant files
Plan — Generate implementation plan from ticket + context
Execute — Run Claude Code in an isolated git worktree
Test — Run test suite, verify no regressions
Self-Review — Automated code review against ERNE standards
Health Delta — Compare audit score before/after
PR — Create pull request with full summary and link to ticket

Supported Providers

clickup — ClickUp tasks (API token + list ID)
github — GitHub Issues (repo + labels)
linear — Linear issues (API key + team)
jira — Jira issues (API token + project)
local — JSON file with ticket definitions (for testing)

Configuration Reference

See worker.example.json at the project root for a full example. Key sections:

provider — Type, credentials, poll interval, filters
repo — Local path, base branch, remote
erne — Hook profile, quality gates, confidence threshold
log — File path and log level

Examples

# Dry run to see available tickets
erne worker --config worker.json --dry-run

# Process one ticket and stop
erne worker --config worker.json --once

# Continuous polling mode
erne worker --config worker.json

Fluxly profile Author and license come from source; runtime, permissions, and network are Fluxly detections or estimates

Fluxly category: Engineering
Author-declared agents: No explicit declaration found; this is not inferred or tested compatibility
Static check: 94 / 100 · heuristic scan, not runtime safety proof
Author / version / license: @JubaKitiashvili · MIT
Fluxly token estimate: Lean
Fluxly setup estimate: Guided setup
External API key: Required · Vendor-specific
Detected OS requirements: Unspecified
Runtime requirements: Unspecified
Detected file/system behavior: Read-only
Write / modify
Shell exec
Detected network behavior: External requests
Install commands: None (reference only)

Profile is derived at build time from SKILL.md and install vectors. Subject to drift from author intent.

Heads up: 未限定 allowed-tools，默认拥有全部工具权限。

Output preview erne-worker.preview

The current SKILL.md does not define a fixed output example.

SKILL.md SKILL.md

# /erne-worker — Autonomous Ticket Execution

## Usage

```bash
erne worker --config <path-to-worker.json>
```

## Options

| Flag | Description |
|------|-------------|
| `--config <path>` | Path to worker configuration JSON (required) |
| `--dry-run` | Fetch tickets and print them without executing |
| `--once` | Process one ticket and exit |

## Pipeline Steps

1. **Poll** — Fetch ready tickets from the configured provider
2. **Validate** — Check ticket has enough detail (title, description, acceptance criteria)
3. **Confidence Score** — Estimate likelihood of autonomous success (0-100)
4. **Context Resolve** — Load project stack, audit data, and relevant files
5. **Plan** — Generate implementation plan from ticket + context
6. **Execute** — Run Claude Code in an isolated git worktree
7. **Test** — Run test suite, verify no regressions
8. **Self-Review** — Automated code review against ERNE standards
9. **Health Delta** — Compare audit score before/after
10. **PR** — Create pull request with full summary and link to ticket

## Supported Providers

- **clickup** — ClickUp tasks (API token + list ID)
- **github** — GitHub Issues (repo + labels)
- **linear** — Linear issues (API key + team)
- **jira** — Jira issues (API token + project)
- **local** — JSON file with ticket definitions (for testing)

## Configuration Reference

See `worker.example.json` at the project root for a full example. Key sections:

- `provider` — Type, credentials, poll interval, filters
- `repo` — Local path, base branch, remote
- `erne` — Hook profile, quality gates, confidence threshold
- `log` — File path and log level

## Examples

```bash
# Dry run to see available tickets
erne worker --config worker.json --dry-run

# Process one ticket and stop
erne worker --config worker.json --once

…

Evidence boundary and execution chain

Author text anchors workflow facts; Fluxly only indexes current sections, terms, files, and commands.

sections -> Usage → Options → Pipeline Steps → Supported Providers → Configuration Reference → Examples
terms    -> Poll · Validate · Confidence Score · Context Resolve · Plan · Execute · Test · Self-Review
files/cmd -> --config <path> · --dry-run · --once · worker.example.json · provider · repo · erne · log
body sha256 -> 8ff2e25b2d88