--- title: "coding.sh workflows · coding.sh docs" description: "Capture, measure and A/B-test multi-agent workflow runs. Every option of coding.sh workflows and its subcommands, from its own --help." canonical: https://coding.sh/docs/cli/workflows human_url: https://coding.sh/docs/cli/workflows markdown_url: https://coding.sh/docs/cli/workflows.md agents_url: https://coding.sh/agents.md llms_url: https://coding.sh/llms.txt updated: 2026-11-11 --- # coding.sh workflows Capture, measure and A/B-test multi-agent workflow runs. Every option of coding.sh workflows and its subcommands, from its own --help. Alias: `workflow, wf`. coding.sh workflows --help ``` coding.sh workflows Capture, measure and A/B-test multi-agent workflow runs Commands list Runs with status, wall time, cost and agent count import Import Claude Code workflow runs from disk show Run tree with per-node time, cost and status stats Critical path, concurrency, barrier idle, failures, cache hits, cost score Grade a run's output against a case's truth run A/B workflow variants over cases; prints the experiment id compare Paired comparison of two variants of an experiment replay Re-run an edited script from a run's journal (cached prefix) Notes Workflow runs are stored under ~/.coding.sh/workflows/ until the index stores them. Aliases workflow, wf ``` ## `coding.sh workflows list` Runs with status, wall time, cost and agent count. coding.sh workflows list --help ``` coding.sh workflows list [options] Runs with status, wall time, cost and agent count Options --framework Only runs from this framework --name Only runs of this workflow --since Only after this time: 7d, 12h, 30m or a date --experiment Only runs of this experiment -n, --limit Maximum rows (default: 20) --import Import new or changed Claude Code workflow runs first (--no-import to skip) (default: true) Examples coding.sh workflows list --since 7d coding.sh workflows list --framework langgraph --json ``` ## `coding.sh workflows import` Import Claude Code workflow runs from disk. coding.sh workflows import --help ``` coding.sh workflows import [path] [options] Import Claude Code workflow runs from disk Arguments path Projects, project, session or run directory (default: Claude Code's projects dir) Options --force Re-import run directories that did not change ``` ## `coding.sh workflows show` Run tree with per-node time, cost and status. coding.sh workflows show --help ``` coding.sh workflows show [options] Run tree with per-node time, cost and status Arguments run Workflow run id Options --tree Include subagent, tool and model-call nodes --gantt Print the timeline as TSV instead ``` ## `coding.sh workflows stats` Critical path, concurrency, barrier idle, failures, cache hits, cost. coding.sh workflows stats --help ``` coding.sh workflows stats Critical path, concurrency, barrier idle, failures, cache hits, cost Arguments run Workflow run id ``` ## `coding.sh workflows score` Grade a run's output against a case's truth. coding.sh workflows score --help ``` coding.sh workflows score [options] Grade a run's output against a case's truth Arguments run Workflow run id Options --case Case id (default: the run's case) --cases Cases directory (default: evals/workflows//cases) --slack Line distance that still matches a seeded bug (default: 5) ``` ## `coding.sh workflows run` A/B workflow variants over cases; prints the experiment id. coding.sh workflows run --help ``` coding.sh workflows run [options] A/B workflow variants over cases; prints the experiment id Arguments workflow Workflow name (cases live under evals/workflows//cases) Options --variant Variant: name=script path (claude), name=command (command), or JSON (repeatable) --cases Cases directory (default: evals/workflows//cases) -k, --k Attempts per case and variant (default: 3) --split Only cases in this split --driver How a variant runs (default: claude) -j, --concurrency Jobs at once (a workflow already fans out) (default: 1) --validate Check seeded bugs on each case's snapshot first: each bug's test fails, and passes with the bug reverted (--no-validate to skip) (default: true) --resume [] Continue an interrupted workflow experiment: finished trials are kept, only the rest run (no ID: the newest unfinished one with the same config) --retries Retries per trial after a rate limit, 5xx or network error, with backoff; a failed setup step is retried once and a refused login never (failures are never retried) (default: 3) --stop-after-infra Stop once N trials in a row fail on infrastructure after their retries (a refused login, a failed setup, rate limits, 5xx, network): no new trial starts and the run is no quality verdict; 0 never stops (default: 5) --trial-timeout Time limit per trial, retries included; a trial over it fails --bin claude binary for the claude-code-workflow driver (default: claude on PATH) -y, --yes Run without asking when the estimated cost is above the confirm threshold Examples coding.sh workflows run coding-sh-improve -v base=wf/improve.js -v cand=wf/improve-v2.js -k 3 --split dev coding.sh workflows run coding-sh-improve -v base=wf/improve.js -v cand=wf/improve-v2.js -k 3 --resume ``` ## `coding.sh workflows compare` Paired comparison of two variants of an experiment. coding.sh workflows compare --help ``` coding.sh workflows compare [options] Paired comparison of two variants of an experiment Arguments experiment Experiment id Options --baseline Baseline variant (required) --candidate Candidate variant (required) --threshold Score that counts as a pass (default: the case's, else 0.5) --seed Bootstrap seed for the intervals and P(front) --select Pick a variant to deploy: cheapest above a quality floor, best quality − λ·cost, or the front's extremes --q-min min-cost: quality lower bound to reach --q-drop min-cost: quality points below the baseline's mean allowed (default 10) --lambda <λ> weighted: quality given up per USD (per second when no run records cost) Examples coding.sh workflows compare wfx_01 --baseline base --candidate cand --select min-cost --q-drop 10 coding.sh workflows compare wfx_01 --baseline base --candidate cand --select pareto --json ``` ## `coding.sh workflows replay` Re-run an edited script from a run's journal (cached prefix). coding.sh workflows replay --help ``` coding.sh workflows replay [options] Re-run an edited script from a run's journal (cached prefix) Arguments run Base workflow run id Options --script Edited workflow script (required) --base-script Script the base run used, when it was not captured --args Workflow args (JSON or a .json path) --dry-run Print the predicted cache hits without running --bin claude binary for the claude-code-workflow driver (default: claude on PATH) -y, --yes Run without asking when the estimated cost is above the confirm threshold ``` [All commands](https://coding.sh/docs/cli) · [Global options](https://coding.sh/docs/cli#global) To set coding.sh up for a user, follow the agent workflow at [/agents.md](https://coding.sh/agents.md).