// Load when a workspace may contain changes from humans or other agents. Covers safe orientation, preserving unfamiliar files, and containing outputs.
| name | shared-workspace |
| type | guardrail |
| description | Load when a workspace may contain changes from humans or other agents. Covers safe orientation, preserving unfamiliar files, and containing outputs. |
| model-invocable | true |
You may be working in a directory that contains human or other-agent work you do not own. Treat unfamiliar files and edits as live work, not clutter.
Isolated worktrees reduce this risk, but they do not remove it: writers, reviewers, prompt editors, and cleanup tasks may still operate in shared directories.
Ownership is not always obvious. After compaction, interruption, or handoff, you may lose memory of files you created earlier. Use the current task, caller instructions, nearby artifacts, and file contents to infer ownership. When ownership is still unclear, preserve the file and report the uncertainty.
Before editing:
Orient to ownership, not cleanliness. A messy workspace may be active work.
When you finish, report:
Load into any agent that receives human direction. Use before acting on instructions, recording decisions, or producing artifacts.
Load when establishing shared vocabulary, resolving term conflicts, disambiguating domain terminology, or auditing consistency across a project's vocab. Shared vocabulary is the contract between human intent and agent action — ambiguity is the root failure mode, vocab files are the record.
Load when reading from, writing to, or maintaining the KB. Covers layer model, structural conventions, and operational tooling.
Use when mining conversation history during dev work — recovering decisions from the top-level primary session, delegating bulk transcript reading to an explorer, or discovering all sessions tied to a work item across interruptions.
Use whenever you need to delegate work to another agent, run tasks in parallel, coordinate multiple agents, or inspect spawn outputs. Also use when routing work to a specific model or provider.
Load before spawning subagents to protect focus and role fit. Make sure to re-read the available agents before spawning a subagent; choose the most specific owner instead of broad defaults, then write a tight handoff with only the context that subagent needs.
