70High
The skill is highly vulnerable to prompt injection and arbitrary code
Behavior Analysis
Description verified — behavior matches claim
Claims to do
Subagent-Driven Development: Execute plan by dispatching fresh subagent per task, with two-stage review after each: spec compliance review first, then code quality review.
Actually does
This skill orchestrates a multi-stage development workflow. It reads an implementation plan, extracts tasks, and manages them in a `TodoWrite` system. For each task, it dispatches an implementer subagent (using `./implementer-prompt.md`), followed by a spec compliance reviewer subagent (using `./spec-reviewer-prompt.md`), and then a code quality reviewer subagent (using `./code-quality-reviewer-prompt.md`). It handles subagent statuses, loops for fixes, and integrates with other 'superpowers' skills like `using-git-worktrees` and `finishing-a-development-branch`.
Skill Info
Nameobra/superpowers/subagent-driven-development
Registrygithub
Version917e5f53b16b
PURLpkg:github/obra/superpowers@917e5f53b16b?skill=subagent-driven-development
Stars180,084
Installs60,600
Source
SHA-256081ad3869e55...
SHA-12ea2f0b09ac8...
MD5543afab876dc...
TLSHd742a8777993...
ssdeep192:FF5kSUXU...
Size12,139 bytes
Install
This skill has been flagged as potentially malicious. Review the findings below before installing.
GitHub
https://github.com/obra/superpowers/tree/main/skills/subagent-driven-development60,600 installs
Claude CodeDocs
/plugin marketplace add obra/superpowers/plugin install subagent-driven-development@obra/superpowers
CodexDocs
Gemini CLIDocs
gemini extensions install https://github.com/obra/superpowers.git --consent
CursorDocs
Skills.shDocs
npx skills add https://github.com/obra/superpowersAssessments (4)
AI Agent Traps ↗External prompt file dependencyhighsupply chain
The skill relies on external Markdown files (e.g., `./implementer-prompt.md`) to define subagent instructions. Compromise of these files could lead to malicious subagent behavior by injecting harmful prompts.
90% confidenceline 55
./implementer-prompt.md, ./spec-reviewer-prompt.md, ./code-quality-reviewer-prompt.md
Subagent code generation and executionmediumcommand execution
Subagents are instructed to implement, test, and commit code, which involves writing files, executing commands, and interacting with version control. While mitigations like git-worktrees are mentioned, this capability could lead to arbitrary code execution or persistence if misused.
80% confidenceline 51
Implementer subagent implements, tests, commits, self-reviews
Multi-stage review may induce approval fatiguemediumapproval fatigue
The iterative two-stage review process, including re-review loops, could lead to human overseers experiencing approval fatigue, potentially overlooking malicious code or actions during the review process.
70% confidenceline 70
Spec reviewer subagent confirms code matches spec? -> Implementer subagent fixes spec gaps -> re-review
Context injection via human inputlowprompt injection
The process explicitly allows for human input to provide missing context to subagents, which could be exploited for prompt injection if the input is not adequately sanitized or validated by the system.
70% confidenceline 50
Implementer subagent asks questions? -> Answer questions, provide context
Version History (4)
Dependencies (1)
superpowers.claude-plugin/marketplace.jsonnot scanned
Badge
Markdown
[](https://mondoo.com/ai-agent-security/skills/github/obra/superpowers/subagent-driven-development)HTML
<a href="https://mondoo.com/ai-agent-security/skills/github/obra/superpowers/subagent-driven-development"><img src="https://mondoo.com/ai-agent-security/api/badge/github/obra/superpowers/subagent-driven-development.svg" alt="Mondoo Skill Check" /></a>Image URL
https://mondoo.com/ai-agent-security/api/badge/github/obra/superpowers/subagent-driven-development.svgSecure your AI agents
Skills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.
- Continuous scanning of skills across all registries
- Policy enforcement before skills reach your agents
- Integration with your existing security workflow