Output Workflow Reset

What it does

When your agent uses it

• Iterating on a later steps prompt
• Logic without re-running the entire workflow
• Recovering from a failure that only affects steps after a known-good point

Example prompts

• “/output-workflow-reset”

Requirements

• Node.js
• Docker
• Pre-approved tools (allowed-tools): Bash

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

1. 1Start from workflow debug — confirm which steps completed before picking a reset point
2. 2Reset to the step whose output you want to keep — reset runs everything after ; the step itself is not re-executed
3. 3Record a --reason — it shows up in Temporal history and helps teammates (and future you) understand why the run forked
4. 4Prefer reset over full rerun for expensive workflows — if early steps make multi-dollar LLM calls, reset saves real money
5. 5Verify side effects first — if an early step writes to an external system, the replay won't re-write, which is usually what you want but…

What it can do on your machine

Read from SKILL.md and the folder at commit 99ee298. It shows what the files ask for, not the result of running them.

• Tool permissions

  Pre-approves these tools, so the agent can use them without asking each time:

  • Bash

  From allowed-tools in the SKILL.md frontmatter.

• Runs code

  Shell commands in SKILL.md call:

  • npx
  • docker
  • curl

  From the folder's file list and the shell code blocks in SKILL.md.

• Network

  No URLs in SKILL.md. Its commands use npx, docker and curl, which can reach the network depending on how they are called.

  From URLs in SKILL.md, links to its own repository left out.

• Credentials

  Names no API keys, tokens, secrets or passwords.

  From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Output Workflow Reset loads about 1.6k tokens when it runs. Until then it costs about 87 tokens; SKILL.md has 659 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it

~87

When it runs · the whole SKILL.md, loaded when a task matches

~1.6k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

• NotePre-approves every shell command (allowed-tools: Bash)SKILL.md

  allowed-tools: Bash

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from growthxai/output at commit 99ee298, republished under its Apache-2.0 licence (© growthxai). 659 words, ~1,583 tokens.

name

output-workflow-reset

description

Re-run an Output SDK workflow from after a specific completed step, creating a new run that replays up to that point and re-executes subsequent steps. Use when iterating on a later step's prompt or logic without re-running the entire workflow, or when recovering from a failure that only affects steps after a known-good point.

allowed-tools

Bash

npx output workflow reset <workflowId> --step <stepName>
npx output workflow reset <workflowId> --step <stepName> --reason "<why>"

npx output workflow debug <workflowId> --json

Workflow reset successfully

Workflow ID: lead_enrichment-a1b2c3d4
New Run ID: 8f3e2a91-...
Reset after step: consolidateCompetitors
Reason: retrying with updated prompt

# The workflow failed at `generateBlogPost`, but `consolidateCompetitors`
# (the step before it) completed successfully.
npx output workflow debug lead_enrichment-a1b2c3d4 --json
# ... confirms consolidateCompetitors completed

# Edit src/workflows/lead_enrichment/prompts/generate_blog_post@v1.prompt
# Then rerun from after the last good step — skipping the expensive
# competitor consolidation LLM call.
npx output workflow reset lead_enrichment-a1b2c3d4 \
  --step consolidateCompetitors \
  --reason "Retry with updated blog-post prompt"

# Workflow completed, but step output is wrong. Rerun just the last step.
npx output workflow reset blog_evaluator-xyz789 --step analyze_claims

# Check the new run's result
npx output workflow result blog_evaluator-xyz789

npx output workflow reset wf-12345 \
  --step fetchCompanyData \
  --reason "Source API returned stale data; rerunning after cache invalidation"

# Grab the new runId from the reset output, then watch it
npx output workflow reset lead_enrichment-a1b2c3d4 --step lookupCompany
npx output workflow status lead_enrichment-a1b2c3d4
npx output workflow result lead_enrichment-a1b2c3d4

© growthxai, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in coding_assistants/claude/plugins/outputai/skills/output-workflow-reset of growthxai/output.

Open the folder on GitHubat commit 99ee298