Skip to content
Agentmart
Submit a tool

issueflow

Umbrella entrypoint for issue-driven Git Flow-lite work.

Skill by jugol

IndexedMIT★ 1
Write a review

Chalk Labs

Need help deploying issueflow?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/jugol-issueflow

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Umbrella entrypoint for issue-driven Git Flow-lite work. Use to adopt issueflow, clarify a feature, continue an autonomous cycle, schedule parallel worktree lanes, govern a plan-first project, or route issue discovery, proof, merge, and compound learning.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/issueflow

What's inside

Skill name
issueflow
Description
Umbrella entrypoint for issue-driven Git Flow-lite work. Use to adopt issueflow, clarify a feature, continue an autonomous cycle, schedule parallel worktree lanes, govern a plan-first project, or route issue discovery, proof, merge, and compound learning.
License
MIT
Instructions
59 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# issueflow

Route the request, preserve project truth, and keep implementation moving through issues to `develop`.

## Hard gates

- In autonomous cycles, keep the main thread as scheduler, reviewer, integrator, and merge-gate owner. Delegate implementation when workers are authorized; otherwise record a no-safe-delegation rationale before editing issue code.
- `issueflow parallel` explicitly authorizes subagents and worktrees. Plain `issueflow` does not. For complex work, shape independent lanes first; request `issueflow parallel` only if missing worker authorization would force main-thread or serialized implementation.
- `Serialized` controls dependency or merge order; it does not authorize main-thread implementation.
- After dispatch, continue read-only scheduling instead of idling. Before `wait_agent`, scan and record Ready/Active, backlog/drafts, plan gaps, proof/test failures, review/feedback findings, solution follow-ups, and stale/deferred candidates. If nothing can dispatch, record one reason: `no-candidate-after-minimum-scan`, `overlaps-active-lane`, `active-lane-budget-full`, or `blocked-on-worker-output`.

## Routing

- New/adopting repo -> `repo-bootstrap`
- Ambiguous feature, product, or UI request -> `issue-brainstorm`
- Concrete problem or issue draft -> `issue-raise`
- Readiness -> `issue-intake`
- Branch/worktree/worker scheduling -> `issue-dispatch`
- Web or Flutter implementation -> `implement-web` or `implement-flutter`
- Proof -> `qa-web-proof` or `qa-flutter-proof`
- `issue/* -> develop` -> `merge-gate`
- Post-merge learning and cleanup -> `issue-compound`
- `develop -> main` -> `release-gate`

## Two tracks

- **Autonomous Cycle:** on continue, iterate, resume, failures, reviews, or feedback, scan durable signals and create the next issue or multi-issue wave without waiting for the user to name every task.
- **Interactive Feature Intake:** read repo context, clarify implementation-critical ambiguity, compare 2-3 approaches, obtain direction, then create an issue or wave. Concrete low-risk bugs may use a compact intake.

If work already changed before routing, backfill the issue, wave, and proof trail honestly.

## Project truth and context

- Read macro then micro context: current-state pointer, plan anchor, current wave/issue, recent proof, branch/worktrees, next action, then matching solution-index entries.
- Treat the active plan as product truth. Classify work as `aligned`, `extension`, `conflict`, or `deviation`; update or record a decision before dispatching extensions/conflicts.
- Plan work must name the target user goal experience and user-visible proof.
- Prefer medium user-visible vertical slices and waves of independent lanes. Combine tiny related findings; do not create one-issue waves without a concrete reason.
- Keep `PLAN_ANCHOR.md` and `CURRENT_STATE.md` bounded. Search archives before opening them; never bulk-load completed issues, old waves/proof, or all solution notes.
- If automation needs user input, credentials, or a product/policy choice, pause it with a resume condition; do not delete it.

## Finish contract

- Start fresh issue work from `develop`; use `issue/*` branches/worktrees for implementation.
- An issue finishes only after merge or PR/merge-queue handoff to `develop`, integrated proof as required, durable state updates, and returning the root checkout to `develop`.
- Record follow-up work as issues or a concrete no-follow-up rationale.
- Run `issue-compound` when work produced reusable learning, prevention rules, failed approaches, follow-up triggers, plan gaps, or stale active context.

## References

Load only what the current decision needs:

- Tracks/planning: `../../references/two-track-routing.md`, `../../references/interactive-brainstorming.md`, `../../references/plan-governance.md`, `../../references/goal-experience-planning.md`
- Scheduling: `../../references/autonomous-wave-generation.md`, `../../references/issue-sizing-and-scheduling.md`, `../../references/parallel-delivery.md`
- Lifecycle/context: `../../references/branch-lifecycle.md`, `../../references/automation-governance.md`, `../../references/compound-learning.md`, `../../references/context-governance.md`, `../../references/history-compaction.md`
- Product shape: `../../references/vertical-slice-architecture.md`, `../../references/experience-first-ui.md`
- Full policy conflicts/onboarding: `../../OPERATING-MODEL.md`
- Durable artifacts: `../../templates/`

From jugol/issueflow-skills, MIT.

More from jugol/issueflow-skills

Signals

Stars
1
Forks
0
Open issues
0
Last commit (this folder)
Jul 11, 2026 (3 months ago)
Last release
Jul 11, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from jugol

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get issueflow running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying issueflow?

CallGet help
Cite this page

issueflow skill. Agent Mart. https://agentmart.dev/skills/jugol-issueflow (last updated 2026-10-24)