agentleFS
MicrosoftKnown publishermicrosoft/PilotSwarm/templates/builder-agents/skills/pilotswarm-agent-versioning/SKILL.md

pilotswarm-agent-versioning

Skill14 starsChanged 8 months agoMITView on GitHub

Use when creating or editing PilotSwarm app agents in agents/*.agent.md. Covers schemaVersion/version frontmatter, version bump guidance, and prompt layer version visibility.

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    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:

    Use this skill when scaffolding or editing app agent definitions for a PilotSwarm-based application.

    line 8
  2. Very specific:

    Every authored .agent.md should include:

    line 12
  3. Very specific:

    Use 3 only when the agent declares initialRequiredTool.

    line 24
  4. Distinctive:

    When creating a new app agent, start with:

    line 30
  5. Very specific:

    When modifying an existing .agent.md, update the version string when the change affects prompt behavior, tool expectations, workflow …

    line 39
  6. Distinctive:

    Use SemVer intent when the app uses SemVer:

    line 41
  7. Distinctive:

    If the existing app uses non-SemVer version labels, preserve that style unless the user asks to migrate.

    line 47
  8. Distinctive:

    When the first turn must execute a real tool before answering, use:

    line 49

≈ “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

PilotSwarm Agent Versioning

Use this skill when scaffolding or editing app agent definitions for a PilotSwarm-based application.

Required Frontmatter

Every authored .agent.md should include:

---
schemaVersion: 1
version: 1.0.0
name: example-agent
description: Example agent.
---
  • schemaVersion is the PilotSwarm agent schema version. Use 1 by default. Use 3 only when the agent declares initialRequiredTool; that tool must also appear in tools. Use 4 when declaring nativeTaskTools: exact application or server/tool names under swarm-explore and/or swarm-task. Parent permissions still apply; durable orchestration and session-control tools cannot be delegated.
  • version is a string. SemVer is recommended because it is easy for operators and agents to compare, but app authors may use another non-empty label.

Creating Agents

When creating a new app agent, start with:

schemaVersion: 1
version: 1.0.0

Editing Agents

When modifying an existing .agent.md, update the version string when the change affects prompt behavior, tool expectations, workflow guidance, frontmatter metadata, output shape, or child-contract expectations.

Use SemVer intent when the app uses SemVer:

  • Patch: wording clarifications or typo fixes that should not materially change behavior.
  • Minor: new capabilities, tools, guidance, examples, or backwards-compatible workflow behavior.
  • Major: changed role semantics, removed expectations, or incompatible output/contract changes.

If the existing app uses non-SemVer version labels, preserve that style unless the user asks to migrate.

When the first turn must execute a real tool before answering, use:

schemaVersion: 3
tools:
	- package_catalog
initialRequiredTool: package_catalog

PilotSwarm corrects one prose-only response, then fails the turn if the tool is still not invoked. Do not simulate this contract with prompt wording alone.

Prompt Layer Visibility

PilotSwarm reports authored prompt layers as:

PilotSwarm base -> App base -> Agent

The Activity Monitor can display session.prompt_layers so operators can see the layer kind, name, type, schema version, and version used by the worker that created or resumed a session.

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.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

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.