agentleFS
Sign inStart free
revealui-studiorevealui-studio/revskills/skills/revealui-ops/SKILL.md

revealui-ops

Skill2 starsChanged 40 days agoMITView on GitHub

Thin shim onto the operational-workflow-layer runner (workflow-run.js). Lists registered workflows (cleanup-session, prepare-for-exit, master-handoff-regen, …) or runs one by name, passing --dry-run/--fix/--yes straight through. Exposed as /ops. One-offs: /cleanup and /rollup. Delegates entirely to the runner — never reimplements workflow execution or safety classification (auto/report-first/gated/owner-only).

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Do NOT invoke a workflow's underlying step commands directly through this skill — always go through workflow-run.js so safety …

    line 59
  2. Very specific:

    Do NOT pass --fix/--yes without telling the user which gated step(s) it will unlock.

    line 60
  3. Very specific:

    Do NOT attempt to run an owner-only step's printed command through another tool on the runner's behalf.

    line 61
  4. Very specific:

    Do NOT treat rfloop / revloop as this runner or as product AgentRuntime.

    line 62

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Thin CLI shim onto the .jv operational-workflow-layer runner (workflow-run.js, design contract at docs/gap-specs/GAP-314-operational-workflow-layer-design.md §4), exposed as /ops. This skill delegates entirely to the runner; it does not reimplement workflow execution, step logic, or safety classification.

rfloop is a PR/CI operator disk state machine only (P0 stub; no LLM; auto-merge locked). It is not the fleet brain or the product AgentRuntime. Prefer rfloop; revloop is a rename shim. This skill is not rfloop.

Load helpers:

. "$REVEALFLEET_ROOT/revskills/scripts/lib/session-state.sh"

/ops or /ops list — discovery

node "$JV_REPO/scripts/workflow-run.js" --list

Prints every registered workflow (name, title, and its declared safety — always the max severity across its steps) from $JV_REPO/workflows/*.yml.

Named one-offs (same runner, do not duplicate logic):

SlashWorkflow
/cleanupcleanup-session
/rollupmaster-handoff-regen

/ops <name> [flags] — run a workflow

node "$JV_REPO/scripts/workflow-run.js" <name> [--dry-run] [--fix] [--yes]

Pass --dry-run, --fix, --yes straight through — the runner defines what each does; this skill does not interpret them. --dry-run prints the plan and executes nothing.

Safety classes (runner-enforced, not this skill)

ClassRunner behavior
autoIdempotent/read-only/report-only. Runs unattended.
report-firstWrites reviewable files (never destructive). Runs and prints; never auto-commits.
gatedDestructive or state-changing. Needs --fix or --yes for that run, or the runner prints STOPPED-GATED and halts the chain there.
owner-onlyDisposition-shaped (per ~/.claude/rules/disposition-actions.md) — merges, gate-clearing labels, remote branch deletes, repo settings. NEVER executed by the runner under any flag; it prints the exact command for the owner and marks SKIPPED-OWNER.

Do not

  • Do NOT invoke a workflow's underlying step commands directly through this skill — always go through workflow-run.js so safety classification applies. The individual commands remain directly invocable outside this skill (that's Tier 2, the escape hatch); this skill is Tier 3, the named-chain path.
  • Do NOT pass --fix/--yes without telling the user which gated step(s) it will unlock.
  • Do NOT attempt to run an owner-only step's printed command through another tool on the runner's behalf. Hand it to the owner verbatim, as printed.
  • Do NOT treat rfloop / revloop as this runner or as product AgentRuntime. rfloop is operator-only (P0 stub; no LLM; auto-merge locked).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

Reviews can't be read right now.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free