Agent skill · Workflow & Productivity

repairing-workspaces-agents

Repairs and upgrades existing Friday workspaces, jobs, user agents, and generated agents to current runtime contracts. Use when a workspace/job/agent is failing, when upgrading old workspace.yml or Python agents, when tools are missing or unauthorized, when jobs emit empty outputs, or when an interactive workflow needs HITL repair.

Friday Platform98★ · 1 repos on radarProfile →
claude-codeNOASSERTION
Install
npx skills add friday-platform/friday-studio --skill repairing-workspaces-agents --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 1
SKILL.md size: 12 KB
Bundled scripts: none
Path: packages/system/skills/repairing-workspaces-agents/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 98
Language: TypeScript
Read our review of the source →

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

From the SKILL.md

# Repairing workspaces and agents Use this skill as the repair runbook for existing Friday workspaces, jobs, and agents. The goal is to preserve the user's intent while upgrading the implementation to the current runtime contracts: host capabilities, refs over data, explicit output emission, and real HITL. Copy this checklist into your working notes: ```text Repair checklist: - [ ] 1. Reproduce or inspect the failing session/job before editing. - [ ] 2. Load companion skills for the surface being repaired. - [ ] 3. Identify the failing boundary: workspace config, signal, job/FSM, agent source, MCP/tool catalog, output contract, or HITL. - [ ] 4. Make the smallest repair that restores the current contract. - [ ] 5. Validate the workspace/job/agent through the daemon, not just by reading files. - [ ] 6. Leave durable notes: what failed, what contract was violated, and how to avoid regenerating the old pattern. ``` ## Load companion skills first Load the narrow skill for the part you are touching: | Repair target | Load | |---|---| | Workspace CRUD, `workspace.yml`, validation, live daemon API | `@friday/workspace-api` | | Jobs, FSM actions, `outputTo`, `inputFrom` | `@friday/writing-

What's inside
Steps it walks through
  1. Load companion skills first
  2. First pass: classify the failure
  3. Current contracts to repair toward
  4. Python/user agents must use host capabilities
  5. outputTo actions must emit output explicitly
  6. HITL is a blocking platform primitive
  7. Tool access requests are only for real tools
  8. Repair workflow
  9. 1. Make a precise diagnosis
  10. 2. Repair the owning boundary
  11. 3. Validate through the daemon
  12. Tool renames — workspaces with hard-coded old names
  13. Gotchas
  14. Repair note template
Commands it runs
grep -RIn 'memory_read\|memory_save\|memory_remove\|workspace_delete\|remove_item\|artifacts_get\|artifacts_create\|get_mcp_dependencies' .
More from friday-studio
All skills →
About this skill
What does the repairing-workspaces-agents skill do?

Repairs and upgrades existing Friday workspaces, jobs, user agents, and generated agents to current runtime contracts. Use when a workspace/job/agent is failing, when upgrading old workspace.yml or Python agents, when tools are missing or unauthorized, when jobs emit empty outputs, or when an interactive workflow needs HITL repair.

How do I install it?

Run `npx skills add friday-platform/friday-studio --skill repairing-workspaces-agents --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From friday-platform/friday-studio, a repository with 98 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going