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 skills add metaspartan/cybara --skill subagent-driven-developmentgit clone --depth 1 https://github.com/metaspartan/cybaraWrote 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/skills/metaspartan/cybara/subagent-driven-development)<a href="https://agentmods.dev/skills/metaspartan/cybara/subagent-driven-development"><img src="https://agentmods.dev/badge/skills/metaspartan/cybara/subagent-driven-development/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/metaspartan/cybara/subagent-driven-development"><img src="https://agentmods.dev/badge/skills/metaspartan/cybara/subagent-driven-development.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 5.1 | $0.00019 | $0.00468 |
| Opus 5 | $0.00010 | $0.00234 |
| Sonnet 5 | $0.00004 | $0.00094 |
| Haiku 4.5 | $0.00002 | $0.00047 |
Grade A, and why
subagent-driven-development 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 9d 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 — 62 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Subagent-Driven Development
Use this for larger implementation plans where independent work can be delegated and reviewed before integration.
This adapts the Hermes/OpenClaw subagent workflow to Cybara's sessions_spawn, sessions_send, sessions_history, session_status, todo, and kanban_* tools.
When To Use
- The plan has multiple mostly independent tasks.
- Fresh context per task reduces confusion.
- You need an implementer plus a spec reviewer and quality reviewer.
- The user explicitly asks for subagents or parallel implementation.
Do not use it for small single-file fixes; direct implementation is faster and clearer.
Workflow
- Read the plan once and extract every task with full context.
- Create a todo list with all task IDs.
- For each task:
- Spawn an implementer session with the exact task, relevant files, constraints, and verification command.
- After implementation, spawn or reuse a reviewer for spec compliance.
- Then run a quality/security/test review.
- Fix gaps and re-review before marking complete.
- After all tasks, run a final integration review across the full diff.
- Run the affected test suite and full project checks required by the repo.
Implementer Prompt Shape
Include:
- Task text copied from the plan.
- Files likely involved.
- Project constraints such as package manager, test command, platform, and style.
- Definition of done.
- A request to avoid unrelated refactors and avoid committing unless instructed.
Reviewer Prompt Shape
Spec reviewer:
- "Compare implementation to this exact task spec."
- "Output PASS or specific gaps with file paths."
Quality reviewer:
- "Check correctness, security, performance, tests, and maintainability."
- "Output Critical, Important, Minor, and Verdict."
Rules
- Never make subagents infer missing requirements from the full plan; pass the relevant task context directly.
- Review receipts count only after both spec and quality review are complete.
- Keep worktree ownership explicit. Do not let parallel agents edit the same files without coordination.
- Final integration review is mandatory for cross-task behavior.
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.
- 9d ago First seen · 62 lines · 19 tokens per session scan A cdb5e7fc07b9
subagent-driven-development is a skill published in the GitHub repository metaspartan/cybara (30 stars, last pushed yesterday), licensed MIT. It adds 19 tokens to every session and 468 once invoked, about $0.0001 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 2026-09-03.
Other skills, from other repositories
kodama-verification
Define measurable success criteria and collect targeted test, build, lint, type-check, or smoke-test evidence before claiming work is complete.
kodama-behavior
Classify requests, assess unfamiliar codebases, delegate work safely, and recover from failed attempts. Use for multi-step, ambiguous, or parallelizable work.
kodama-constraints
Enforce non-negotiable safety, scope, security, and quality constraints for implementation and review work.
project-release-check
Validate this example project's release readiness. Use before publishing a release candidate.
slop-taste
Anti-slop frontend skill for landing pages, portfolios, and redesigns. The agent reads the brief, infers the right design direction, and ships interfaces that do not look templated. Real design systems when applicable, audit-first on redesigns, strict pre-flight check.
slop-hallmark
Anti-AI-slop design skill for greenfield pages, audits, redesigns, and design extraction from URLs or screenshots. Use when the user asks to build a new app or landing page, wants to redesign something, invokes Hallmark by name, or uses audit/redesign/study.