v7.0.0

RePrompter

ReviewClawScan verdict for this skill. Analyzed May 1, 2026, 5:45 AM.

Analysis

RePrompter is a coherent prompt-improvement skill, but its team mode can launch tmux-based agent teams that execute and retry work, so it deserves careful review before installation.

GuidanceInstall this only if you want both prompt cleanup and optional team orchestration. For normal prompt rewriting, use Single mode. Before using Repromptception, smart run, or run-with-quality modes, inspect the generated plan, run in a version-controlled workspace, monitor tmux sessions, and clean up temporary prompt/output files.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation

SeverityMediumConfidenceHighStatusConcern

SKILL.md

Repromptception ... Plan team → reprompt each agent → tmux Agent Teams → evaluate → retry ... Direct code execution belongs to coding-agent unless Repromptception execution mode is explicitly requested.

The skill can move from prompt cleanup into tmux-based multi-agent execution with retries when Repromptception execution is requested.

User impactIf invoked in team/run mode, the agent may coordinate multiple execution agents rather than only rewrite a prompt.

RecommendationUse Single mode for prompt cleanup. Only use Repromptception/run modes after inspecting the generated team brief, confirming the intended scope, and running in a safe project state.

Rogue Agents

SeverityMediumConfidenceHighStatusConcern

TESTING.md

Phase 3: tmux session created, agents execute in parallel

The documented workflow creates a persistent tmux session for parallel agents, which is a higher-impact execution pattern than a one-shot prompt transformation.

User impactParallel agent sessions may continue consuming attention, context, or local execution capability until the user monitors and stops them.

RecommendationBefore using team mode, know how to inspect and terminate the tmux session, and prefer running it in a clean workspace with version control.

Agentic Supply Chain Vulnerabilities

SeverityLowConfidenceHighStatusNote

SKILL.md

Repromptception mode requires Claude Code with tmux and CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1.

The skill discloses a runtime dependency and experimental setting in SKILL.md, while the registry requirements list no required binaries or environment variables.

User impactUsers may not realize team mode depends on tmux and an experimental Agent Teams setting until they read the skill instructions.

RecommendationDocument tmux and the Agent Teams environment variable in registry metadata or installation guidance.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Memory and Context Poisoning

SeverityLowConfidenceHighStatusNote

TESTING.md

Team brief written to `/tmp/rpt-brief-*.md` ... Per-agent XML prompts written to `/tmp/rpt-agent-prompts-*.md`

Generated team briefs and per-agent prompts can persist as local files and may contain project or task context.

User impactSensitive project details included in generated prompts could remain in temporary files after a run.

RecommendationReview generated files before sharing them, avoid including secrets in prompts, and delete temporary output files when no longer needed.