continue-work
Unblocks stalled development tasks by resetting loop budgets and relaunching implementation sessions.
Install
mkdir -p .claude/skills/continue-work && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/16458" && unzip -o skill.zip -d .claude/skills/continue-work && rm skill.zipInstalls to .claude/skills/continue-work
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Resumes a stalled task by resetting its remediation loop counter and relaunching a `code` agent. A task becomes `stalled` when the code skill exhausts its loop budget (default 10 iterations) without satisfying all Definition of Done criteria. This skill shows the user the last failing criteria, lets them add optional guidance, resets the counter (with an optional new budget), and triggers a fresh implementation session. Optionally accepts `--max-loops N` to set a custom loop budget for the new session. Trigger on: "continue-work TASK-NNN", "/continue-work", "resume TASK-NNN", "retry TASK-NNN", "unblock TASK-NNN", "restart work on TASK-NNN", "reset loops for TASK-NNN", "give TASK-NNN more loops", or any request to resume a task that is in `stalled` status.Key capabilities
- →Reset the remediation loop counter for a stalled task
- →Relaunch a fresh code session with the original context
- →Accept optional user guidance for the new session
- →Set a custom loop budget for the new session
- →Update task status to 'todo' and notify the board
How it works
This skill locates a stalled task, presents its stall context to the user, and then resets the task's loop counters and status. It then relaunches the code skill for the task with the original context and any new guidance.
Inputs & outputs
When to use continue-work
- →Resume a task that hit a loop limit
- →Retry an automated coding task with new instructions
- →Unblock a stalled development workflow
About continue-work
Resets the task loop counter for stalled processes, accepts optional user feedback, and triggers a fresh code generation session to satisfy the original Definition of Done.
>
When not to use it
- →The task status is 'done'
- →The task status is 'in_progress'
- →The task status is 'in_review'
Limitations
- →This skill only resumes 'stalled' tasks.
- →This skill does not modify design decisions or task definitions.
- →This skill does not create new branches.
How it compares
This skill automates the process of restarting a stalled task, including resetting counters and re-injecting context, which would otherwise require manual file edits and command invocations.
Compared to similar skills
continue-work side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| continue-work (this skill) | 0 | 4mo | Review | Beginner |
| cursor-local-dev-loop | 3 | 2mo | Review | Beginner |
| codex-worker | 1 | 7mo | Review | Advanced |
| agent-challenges | 1 | 7mo | No flags | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
cursor-local-dev-loop
jeremylongshore
Optimize local development workflow with Cursor. Triggers on "cursor workflow", "cursor development loop", "cursor productivity", "cursor daily workflow". Use when working with cursor local dev loop functionality. Trigger with phrases like "cursor local dev loop", "cursor loop", "cursor".
codex-worker
MoonshotAI
Spawn and manage multiple Codex CLI agents via tmux to work on tasks in parallel. Use whenever a task can be decomposed into independent subtasks (e.g. batch triage, parallel fixes, multi-file refactors). When codex and tmux are available, prefer this over the built-in Task tool for parallelism.
agent-challenges
ruvnet
Agent skill for challenges - invoke with $agent-challenges
sensei
Azure
**WORKFLOW SKILL** — Iteratively improve skill frontmatter compliance using the Ralph loop pattern. WHEN: "run sensei", "sensei help", "improve skill", "fix frontmatter", "skill compliance", "frontmatter audit", "score skill", "check skill tokens". INVOKES: token counting tools, test runners, git co
workflow-update
nomadoor
Use when editing example_workflows JSON files or changing node schema in ways that affect sample workflows.
gsd-fast
Esteban-Marketing
Execute a trivial task inline — no subagents, no planning overhead