Skills Agent Coordination

📦

Agent Coordination

Name: Agent Coordination
Author: ClementWalter

Content revision r2 High Risk ⚙️ External commands🔑 Env variables🌐 Network access📁 Filesystem access

Coordinate coding agents with clear tasks

Multi-agent work can lose context, ownership, and verification. This skill provides delegation patterns, task templates, VibeKanban procedures, and CI monitoring guidance.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Stop for confirmation before installing.

Review the plan and obtain explicit user consent before changing files.

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 Coordination" from https://skillstore.io/skills/clementwalter-agent-coordination.md and its manifest at https://skillstore.io/api/skills/clementwalter-agent-coordination/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

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/clementwalter-agent-coordination.md Signed manifest GET /api/skills/clementwalter-agent-coordination/manifest Signed lockfile GET /api/skills/clementwalter-agent-coordination/lockfile

Test it

Using "Agent Coordination". Create a task to fix an intermittent login failure.

Expected outcome:

Title: Bug: Investigate intermittent login failure
Acceptance criteria: reproduce the failure, identify the cause, add regression coverage, and verify the fix.
Out of scope: redesigning the login interface.

Using "Agent Coordination". Can these tasks run at the same time? One changes authentication and one changes billing.

Expected outcome:

Recommendation: Run in parallel if the tasks do not share configuration or common files.
Verification: each agent reports changed files and test results before integration.

Security Audit

High Risk

v9 • 7/18/2026 Open versioned report

All 60 static findings are false positives: they match Markdown formatting, sample task documentation, API references, or a constrained GitHub CLI CI monitor. The review identified two intent-level concerns in SKILL.md: persistent role instructions and an unconditional local autonomy exception that can weaken normal user control.

Files scanned

1,647

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (2)

High

SKILL.md:14-21 SKILL.md:23-27

Persistent coordinator-mode instruction

The skill instructs the assistant that a CoS mode persists for the entire conversation and that all subsequent requests must become task delegations. This can override a user’s later request context and is a prompt-injection-style behavioral constraint.

The cited instructions explicitly require persistent behavior and redefine all later user requests. Although related to the skill purpose, the unconditional persistence creates a material control-flow risk.

High

SKILL.md:50-56

Unbounded local autonomy exception

The skill grants full autonomy for a named local path without requiring confirmation. A path-based exception can bypass normal user-approval safeguards for operations in that repository.

The exception explicitly removes the coordinator restriction for a specific local directory. It does not define operation limits or a confirmation requirement.

Risk Factors

⚙️ External commands (30)

🔑 Env variables (2)

references/delegation-patterns.md:165 references/delegation-patterns.md:196

🌐 Network access (6)

references/vibekanban-api.md:221 scripts/monitor-pr-ci.py:16 scripts/monitor-pr-ci.py:71 scripts/monitor-pr-ci.py:78 scripts/monitor-pr-ci.py:79 scripts/monitor-pr-ci.py:139

📁 Filesystem access (1)

references/vibekanban-api.md:139

Audited by: claude 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

Copy report link

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

Markdown badge

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

HTML badge

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

Embed card

<iframe src="https://skillstore.io/embed/skills/clementwalter-agent-coordination.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

ClementWalter. (2026). Agent Coordination security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/clementwalter-agent-coordination/audits/9

BibTeX citation

@techreport{clementwalter-clementwalter-agent-coordination-2026,
  author = {ClementWalter},
  title = {Agent Coordination security audit report (audit version 9)},
  institution = {Skillstore},
  year = {2026},
  number = {9},
  url = {https://skillstore.io/skills/clementwalter-agent-coordination/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 Coordination security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
  - name: "ClementWalter"
date-released: "2026-07-18"
url: "https://skillstore.io/skills/clementwalter-agent-coordination/audits/9"
identifiers:
  - type: other
    value: "skillstore:clementwalter-agent-coordination:audit:9"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Break down a feature for an agent team

Turn a feature request into an investigation task and focused implementation tasks with acceptance criteria.

Manage parallel maintenance work

Identify independent directory-level work and sequence tasks that share files or configuration.

Track pull request CI

Check a pull request status and review failed check output with the included monitoring utility.

Try These Prompts

Create a first task

Act as a coordinator. Create a task for adding dark mode. Include the goal, context, acceptance criteria, and out-of-scope items.

Plan agent delegation

Break this feature into investigation and implementation tasks. Identify dependencies and assign clear verification criteria: [feature description].

Choose execution order

Review these proposed tasks. Mark each as parallel or sequential, explain shared-state conflicts, and give the recommended order: [task list].

Run a coordination review

Act as chief of staff for this project. Summarize active tasks, blockers, decisions needed, and the next delegation actions: [project context].

Best Practices

Give every task a measurable acceptance criterion and an explicit scope boundary.
Use an investigation task before implementation when the scope or risk is unclear.
Confirm repository permissions before retrieving CI details or logs.

Avoid

Do not create vague tasks that only say to fix or improve something.
Do not run dependent tasks in parallel when they modify shared files or configuration.
Do not treat CI output as safe to share because failed logs can contain sensitive data.