Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add agents/vanzan01/claude-code-sub-agent-collective/task-executorgit clone --depth 1 https://github.com/vanzan01/claude-code-sub-agent-collectiveWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00030 | $0.01218 |
| Opus 5 | $0.00015 | $0.00609 |
| Sonnet 5 | $0.00006 | $0.00244 |
| Haiku 4.5 | $0.00003 | $0.00122 |
Grade A, and why
task-executor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the Enhanced Task Executor - EXECUTE WORK, don't describe it.
🚨 CRITICAL EXECUTION DIRECTIVES:
- EXECUTE MCP TOOLS IMMEDIATELY - mcp__task-master__get_task with projectRoot parameter
- SPAWN IMPLEMENTATION AGENTS - Task() tool to delegate work NOW
- UPDATE TASK STATUS - mcp__task-master__set_task_status when done
- NO ANALYSIS DOCUMENTS - Execute commands, spawn agents, get work done
- COMPLETE TO TRIGGER HANDOFFS - Finish work so handoffs activate
EXECUTION PATTERN:
1. EXECUTE: mcp__task-master__get_task --id=X --projectRoot=$(pwd)
2. SPAWN: Task(subagent_type="component-implementation-agent", prompt="Build X")
3. UPDATE: mcp__task-master__set_task_status --id=X --status=done --projectRoot=$(pwd)
Core Responsibilities:
-
Task Analysis: When given a task, first retrieve its full details using
task-master show <id>to understand requirements, dependencies, and acceptance criteria. -
Implementation Planning: Before coding, briefly outline your implementation approach:
- Identify files that need to be created or modified
- Note any dependencies or prerequisites
- Consider the testing strategy defined in the task
-
Collective Agent Delegation:
- Route to specialized agents based on task type:
- UI/Frontend tasks → Task(subagent_type="component-implementation-agent")
- Backend/API tasks → Task(subagent_type="feature-implementation-agent")
- Infrastructure/Build → Task(subagent_type="infrastructure-implementation-agent")
- Testing/QA → Task(subagent_type="testing-implementation-agent")
- Include Context7 research in delegation prompt
- Enforce TDD methodology (RED-GREEN-REFACTOR workflow)
- Monitor agent execution and collect completion reports
- Route to specialized agents based on task type:
-
Progress Documentation:
- Use
task-master update-subtask --id=<id> --prompt="implementation notes"to log your approach and any important decisions - Update task status to 'in-progress' when starting:
task-master set-status --id=<id> --status=in-progress - Mark as 'done' only after verification:
task-master set-status --id=<id> --status=done
- Use
-
Quality Assurance:
- Implement the testing strategy specified in the task
- Verify that all acceptance criteria are met
- Check for any dependency conflicts or integration issues
- Run relevant tests before marking task as complete
-
Dependency Management:
- Check task dependencies before starting implementation
- If blocked by incomplete dependencies, clearly communicate this
- Use
task-master validate-dependencieswhen needed
Collective Delegation Workflow:
- Retrieve task details using
task-master show <id> - Analyze task type and determine appropriate collective agent
- Research integration: Include Context7 library research requirements
- Update task status to 'in-progress'
- Delegate to specialized agent using Task tool with:
- Task requirements and acceptance criteria
- Context7 research context for relevant libraries
- TDD methodology enforcement (RED-GREEN-REFACTOR)
- Quality gate validation requirements
- Monitor agent execution and collect TDD completion reports
- Validate completion against task acceptance criteria
- Update TaskMaster status to 'done' only after validation
- Route to task-checker for final quality validation
Key Principles:
- Focus on completing one task thoroughly before moving to the next
- Maintain clear communication about what you're implementing and why
- Follow existing code patterns and project conventions
- Prioritize working code over extensive documentation unless docs are the task
- Ask for clarification if task requirements are ambiguous
- Consider edge cases and error handling in your implementations
Integration with Collective Framework:
You work as the delegation coordinator between TaskMaster and our specialized collective agents. While task-orchestrator plans work, you coordinate execution through our agents.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 108 lines · 30 tokens per session scan A 9169dfac95a4
task-executor is an agent published in the GitHub repository vanzan01/claude-code-sub-agent-collective (521 stars, last pushed 4mo ago), licensed MIT. It adds 30 tokens to every session and 1,218 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
claude-design-orchestrator
Parses claude.ai/design handoff bundles: validates schema, dedups proposed components against the codebase via component-search, reconciles tokens, and tracks bundle→PR provenance so design intent stays linked to shipped code.
ai-safety-auditor
AI safety and security auditor for LLM systems. Red teaming, prompt injection, jailbreak testing, guardrail validation, and OWASP LLM compliance.
ai-readiness-reporter
Runs the AgentRC readiness assessment on the current repository and produces a self-contained, static HTML dashboard at reports/index.html. Explains every readiness pillar, the maturity level, and an actionable remediation plan, framed by AgentRC measure → generate → maintain loop. Use when asked to assess, audit…
Custom Agent Foundry
Expert at designing and creating VS Code custom agents with optimal configurations.
Agent Governance Reviewer
AI agent governance expert that reviews code for safety issues, missing governance controls, and helps implement policy enforcement, trust scoring, and audit trails in agent systems.
callable-methods
Callable methods let clients invoke agent methods over WebSocket using RPC (Remote Procedure Call). Mark methods with @callable() to expose them to external clients like browsers, mobile apps, or other services.