Is subagent-driven-development safe?
Grade A, 100/100. No meaningful risk found.
Use when executing implementation plans with independent tasks in the current session
Asubagent-driven-development
skills/subagent-driven-development100/100 · No meaningful risk foundNo security findings
subagent-driven-development
skills/subagent-driven-developmentUse when executing implementation plans with independent tasks in the current session
- Content hash
- sha256:c985e5fbf287de92660b825116caa7106a6d598a10138348eec922264e128845
- Files
- 7
- Quality
- 100/100
Security findings
None of the 30 rules matched.
Quality
- infoSKILL.md is longSL705 · skills/subagent-driven-development/SKILL.md:1
569 linesLong skills spend context on every load. Fix: Keep SKILL.md under 500 lines and move reference material into linked files.
Files in this skill
- SKILL.md31.8 KB
- implementer-prompt.md6.2 KB
- re-review-prompt.md4.6 KB
- task-reviewer-prompt.md8.8 KB
- scripts/review-package2.0 KB
- scripts/sdd-workspace3.6 KB
- scripts/task-brief1.3 KB
Questions about subagent-driven-development
- Is subagent-driven-development safe to install?
- Our scanner gave subagent-driven-development grade A (100/100): no meaningful risk found. None of the 30 security rules matched. Static analysis cannot prove safety, so pin the reviewed revision with a lockfile.
- How do I install subagent-driven-development in Claude Code or Codex?
- Run skillhash add github:obra/superpowers/skills/subagent-driven-development. It downloads the exact commit, checks the grade against your policy, writes skills.lock and installs it into .claude/skills and .agents/skills.
- What happens if subagent-driven-development changes upstream?
- Its content hash changes. skillhash verify fails in CI until you review the new version and update the lock. With a watch, you also get an alert.
More from obra/superpowers
- A
requesting-code-review
302kobra/superpowers
Use when completing tasks, implementing major features, or before merging to verify work meets requirements
No security findings - A
writing-plans
302kobra/superpowers
Use when you have a spec or requirements for a multi-step task, before touching code
No security findings - A
writing-skills
302kobra/superpowers
Use when creating new skills, editing existing skills, or verifying skills work before deployment
No security findings - A
diagnosing-superpowers
302kobra/superpowers
Use when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.
No security findings - A
dispatching-parallel-agents
302kobra/superpowers
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
No security findings - A
executing-plans
302kobra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
No security findings - A
finishing-a-development-branch
302kobra/superpowers
Use when implementation is complete, all tests pass, and you need to decide how to integrate the work
No security findings - A
receiving-code-review
302kobra/superpowers
Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation
No security findings