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/stepanenkoviktor0110-boop/ai-dev-methodology-codex/task-validatorgit clone --depth 1 https://github.com/stepanenkoviktor0110-boop/ai-dev-methodology-codexWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/stepanenkoviktor0110-boop/ai-dev-methodology-codex/task-validator)<a href="https://agentmods.dev/agents/stepanenkoviktor0110-boop/ai-dev-methodology-codex/task-validator"><img src="https://agentmods.dev/badge/agents/stepanenkoviktor0110-boop/ai-dev-methodology-codex/task-validator.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00063 | $0.02786 |
| Opus 5 | $0.00032 | $0.01393 |
| Sonnet 5 | $0.00013 | $0.00557 |
| Haiku 4.5 | $0.00006 | $0.00279 |
Grade A, and why
task-validator 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 4d 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.
The source is not reproduced here
No licence file
A repository with no LICENSE is all rights reserved by default, so the body is not copied here. The metadata, the measurements and the link are.
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.
- 4d ago First seen · 220 lines · 63 tokens per session scan A 2fc2a67237e7
task-validator is an agent published in the GitHub repository stepanenkoviktor0110-boop/ai-dev-methodology-codex (1 stars, last pushed 4mo ago), with no licence file. It adds 63 tokens to every session and 2,786 once invoked, about $0.0003 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 2026-08-31.
Other agents, from other repositories
joycraft-researcher
Independent research agent — sees only questions, never the brief.
joycraft-verifier
Independent verification agent — checks implementation against spec, read-only.
pr-reviewer
AWK PR Reviewer. Executes complete PR review flow: prepare -> review implementation -> verify tests -> submit. Used when analyze-next returns reviewpr.
conflict-resolver
AWK Merge Conflict Resolver. Resolves git merge conflicts in a worktree. Used when dispatch-worker returns needsconflictresolution status.
architecture-reviewer
AWK Architecture Reviewer. Performs architecture-level review focusing on code organization, module boundaries, dependency direction, and design patterns. Used as optional second reviewer when review.multimodel is enabled.
worker
执行工（士兵），用于挂帅/orchestrator 模式——主会话（STRONG 模型）把已经拆解清楚、边界明确的子任务派给它并行执行：写样板代码、批量改文件、跑测试、常规检索等。军令必须自包含（含文件路径与验收标准）。.