Skills agent-fungibility
📦

agent-fungibility

Content revision r1 Safe ⚙️ External commands

Coordinate Fungible Agent Swarms

Multi-agent software projects can stall when agents have fixed roles and fragile handoffs. This skill teaches a shared-task workflow where any Claude, Codex, or Claude Code agent can take the next ready bead.

Supports: Claude Codex Code(CC)
🥉 78 Bronze

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "agent-fungibility" from https://skillstore.io/skills/dicklesworthstone-agent-fungibility.md and its manifest at https://skillstore.io/api/skills/dicklesworthstone-agent-fungibility/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/dicklesworthstone-agent-fungibility.md Signed manifest GET /api/skills/dicklesworthstone-agent-fungibility/manifest Signed lockfile GET /api/skills/dicklesworthstone-agent-fungibility/lockfile

Test it

Using "agent-fungibility". I need to start five agents on a backlog with mixed coding and testing work.

Expected outcome:

Use one shared starting context, convert the plan into dependency-aware beads, and let every agent choose the next ready bead. Add agents without permanent roles.

Using "agent-fungibility". One Claude Code session failed during a refactor.

Expected outcome:

  • Treat bead state as the source of truth.
  • Start a replacement session with the same project context.
  • Have the new agent inspect active beads and either resume or choose ready work.

Using "agent-fungibility". Our team wants separate tester, backend, and reviewer agents.

Expected outcome:

Use specialized roles only when the role structure creates value. For normal software delivery, keep agents interchangeable and encode needs in each bead.

Security Audit

Safe
v9 • 7/6/2026 Open versioned report

All six static findings are false positives from Markdown inline code and fenced command examples. The skill describes manual BV and NTM commands, but it does not execute code, use dynamic shell input, or show malicious intent. No evidence found for prompt injection, data exfiltration, or business-logic abuse in SKILL.md.

1
Files scanned
199
Lines analyzed
0
Review items
0
False positives ignored

Risk Factors

⚙️ External commands (6)
SKILL.md:58 SKILL.md:152-158 SKILL.md:158-162 SKILL.md:162-170 SKILL.md:157 SKILL.md:152-158
No confirmed security findings were detected by the latest completed static and semantic audit. This does not prove the skill has no side effects.
Audited by: codex View Audit History →
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/dicklesworthstone-agent-fungibility/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/dicklesworthstone-agent-fungibility/security.svg)](https://skillstore.io/skills/dicklesworthstone-agent-fungibility?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/dicklesworthstone-agent-fungibility?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/dicklesworthstone-agent-fungibility/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/dicklesworthstone-agent-fungibility.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA · BibTeX · CFF)

APA citation

Dicklesworthstone. (2026). agent-fungibility security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/dicklesworthstone-agent-fungibility/audits/9

BibTeX citation

@techreport{dicklesworthstone-dicklesworthstone-agent-fungibility-2026, author = {Dicklesworthstone}, title = {agent-fungibility security audit report (audit version 9)}, institution = {Skillstore}, year = {2026}, number = {9}, url = {https://skillstore.io/skills/dicklesworthstone-agent-fungibility/audits/9}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "agent-fungibility security audit report (audit version 9)" version: "unspecified" type: report authors: - name: "Dicklesworthstone" date-released: "2026-07-06" url: "https://skillstore.io/skills/dicklesworthstone-agent-fungibility/audits/9" identifiers: - type: other value: "skillstore:dicklesworthstone-agent-fungibility:audit:9" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: High
55
Architecture
85
Maintainability
87
Content
69
Community
83
Spec Compliance

What You Can Build

Scale a Coding Swarm

Turn a planned backlog into ready beads so many agents can work without permanent role assignments.

Recover Failed Sessions

Replace a crashed or confused agent by using shared project context and current bead state.

Evaluate Coordination Models

Compare fungible swarms with specialized roles before designing a multi-agent development workflow.

Try These Prompts

Explain the Model 
Explain agent fungibility for a software team using Claude, Codex, and Claude Code. Include the main benefits and when not to use it.
Plan a Small Swarm 
Create a starter workflow for three fungible agents using beads, BV triage, and Agent Mail. Keep setup steps concise.
Recover a Stalled Project 
Review this agent swarm status and propose how to replace failed agents without role mapping. Focus on bead state and shared context: [status].
Design a Large Swarm Protocol 
Design a scalable fungible-agent operating model for [project]. Include task intake, readiness rules, communication norms, failure recovery, and governance checks.

Best Practices

  • Put project context, constraints, and acceptance criteria into the plan before creating beads.
  • Use shared readiness rules so every agent can choose work without central assignment.
  • Keep communication brief and make state changes visible in beads and agent mail.

Avoid

  • Assigning permanent specialist roles before the backlog proves they are needed.
  • Letting agents wait for handoffs when a ready bead is available.
  • Using a shared initial prompt without reviewing commands for the current repository.

Frequently Asked Questions

What is agent fungibility?
It means each agent can handle any ready task instead of holding a permanent specialist role.
Which tools does this skill mention?
It mentions Claude Code, Codex, Gemini CLI, Amp, Cursor, BV, NTM, and Agent Mail.
Does this skill install or run automation?
No. It is guidance and command examples for a workflow that the user controls.
When are specialized agents better?
They can help when structured debate or distinct perspectives are the main purpose of the system.
Can it work with mixed agent products?
Yes. The skill describes using different agent products as interchangeable workers on the same bead queue.
What should I review before use?
Review local BV, NTM, Agent Mail setup, command examples, repository permissions, and any data shared with agents.

Developer Details

License

MIT

Skillstore revision

r1

Version notice

The author did not declare a version.

Repository

https://github.com/Dicklesworthstone/agent_flywheel_clawdbot_skills_and_integrations/tree/main/skills/agent-fungibility

Ref

72d5025b022c77f7a51bdf5c1637c689c80e89d1

Maintenance freshness

7/18/2026

Usage

7 downloads · 149 views

File structure

📄 SKILL.md