apra-labs/apra-fleet/auto-sprint-args
Claude Code and agent skill by apra-labs/apra-fleet
- community
Checking that the SKILL.md file is still in the repository.
How to construct and pass correct arguments to the auto-sprint workflow -- the required issues+branch fields, optional goal/max_cycles/base_branch/requirementsFile, JSON-object (not string) shape, and common launch mistakes. Trigger when preparing to launch or invoke auto-sprint.
Install
npx vskill@latest install apra-labs/apra-fleet/auto-sprint-argsSKILL.md
auto-sprint argument contract
Auto-sprint hard-fails on malformed args (validateSprintArgs runs first, before any work). Get the shape right before launching.
Fields
| Field | Required | Type | Default | Notes | |
|---|---|---|---|---|---|
issues | yes | array of strings | -- | Beads issue IDs (sprint roots), e.g. ["BD-1","BD-2"]. Must be non-empty; every entry a non-empty string. | |
branch | yes | string | -- | Sprint branch name, e.g. "feat/auth". Created from origin/<base_branch> if it does not exist. Must differ from base_branch. | |
goal | no | string | "P1/P2" | Exit when no open issues at or above this priority. Must be exactly "P1", "P1/P2", or "P1/P2/P3". | |
max_cycles | no | positive integer | 5 | Hard cycle ceiling. | |
base_branch | no | string | "main" | PR target branch. | |
requirementsFile | no | string | none | Path to an additional context file for the planner. | |
skip_dolt_push | no | boolean | false | When true, skip the Harvest bd dolt push step so the sprint never writes to a real Dolt remote. Set this in CI/e2e runs. |
Do not invent other fields -- these seven are the whole contract.
Args must be a JSON OBJECT, not a JSON-encoded string
RIGHT -- pass the object itself as the args value:
Full example:
WRONG -- a JSON string containing escaped JSON:
Also wrong: { "issues": "APM-12" } (issues must be an array), { "issue": [...] } (field is issues), { "goal": "P2" } (not an accepted goal value).
Tolerant fallback forms (accepted, but don't rely on them)
The parser normalizes these to { issues: [...] }:
BD-1-- bare issue IDBD-1 BD-2orBD-1,BD-2-- space/comma-separated IDs["BD-1","BD-2"]-- JSON array
In these forms branch falls back to the current git branch -- an easy way to accidentally sprint on the wrong branch. Always use the full object form with an explicit branch.
Pre-launch checklist (mirrors the preflight gate)
- Issues exist and are open --
bd show <id>for each root; a missing root hard-fails preflight. - Ready leaf work exists -- at least some leaves under the roots are unblocked (
bd ready); an all-blocked backlog deadlocks and aborts. - Branch name chosen -- explicit
branch, created if absent. - `base_branch` is the intended PR target (default
main) and must differ frombranch-- an equal pair is rejected at launch. - Branching off latest main -- the workflow fetches and cuts new branches from
origin/<base_branch>; if fetch fails, launch fails. Make sure the remote is reachable.
Common failures and the exact error you'll see
| Mistake | Error emitted | |
|---|---|---|
| Args passed as escaped-JSON string or wrong shape | invalid args: issues -- "issues" must be a non-empty array of beads IDs. Expected a JSON OBJECT, e.g. {"issues":["BD-7"],"branch":"feat/x"}. Received: <your raw args> | |
| No issues at all | invalid args: issues -- "issues" must be a non-empty array of beads IDs | |
Non-string entries in issues | invalid args: issues entries | |
Bad goal / max_cycles / branch / base_branch value | invalid args: <field> with the expected type/values | |
| Root issue doesn't exist | preflight: root <id> not found | |
| All roots already closed | Clean exit: all sprint roots already closed (not an error) | |
| Open issues but every leaf blocked | deadlock: open issues but none ready -- with a per-leaf blocked_by diagnostic | |
git fetch fails (stale main risk) | preflight: git fetch failed -- cannot guarantee branch is off latest <base_branch> | |
| Another run live on the same branch | preflight: another auto-sprint run appears active on <branch> (state age <n>s < TTL) |
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingpackages/apra-fleet-se/apra-pm/.claude/skills/auto-sprint-args/SKILL.md
Popularity
- Repository stars
- 104
- Repository forks
- 18
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/apra-labs/apra-fleet/auto-sprint-args)