Security audit
Security checks for vulnerabilities and agentic risk
This is a planning-review skill that inspects the repository and may write planning documents, but the behavior is mostly coherent with its stated purpose and does not show deception, exfiltration, or destructive actions.
Install this only if you want an opinionated, repository-aware planning reviewer. Expect it to inspect git and project documents, ask multiple decision questions, create local planning notes in expansion modes, and possibly use a reviewer subagent; review generated TODO or docs changes before accepting them.
The trigger conditions are broad phrases like 'think bigger' or whenever a plan 'could be thinking bigger,' which makes accidental invocation likely. Because this skill performs repository inspection, asks repeated decision questions, and may write artifacts, over-broad activation can unexpectedly steer workflows and consume authority in contexts where the user did not request it.
The instruction to 'always prefer the complete option over shortcuts' biases the agent's recommendations in a predetermined direction rather than presenting balanced choices. In context, this can systematically nudge users toward larger, riskier, or more invasive changes, especially when the same skill can inspect repos and propose follow-on modifications.
**ALWAYS follow this structure for every question:**
1. **Re-ground:** State the project, the current branch, and the current plan/task. (1-2 sentences)
2. **Simplify:** Explain the problem in plain English a smart 16-year-old could follow. No raw function names, no internal jargon, no implementation details. Use concrete examples and analogies. Say what it DOES, not what it's called.
3. **Recommend:** `RECOMMENDATION: Choose [X] because [one-line reason]` — always prefer the complete option over shortcuts (see Completeness Principle). Include `Completeness: X/10` for each option. Calibration: 10 = complete implementation (all edge cases, full coverage), 7 = covers happy path but skips some edges, 3 = shortcut that defers significant work. If both options are 8+, pick the higher; if one is ≤5, flag it.
4. **Options:** Lettered options: `A) ... B) ... C) ...` — when an option involves effort, show both scales.
Assume the user hasn't looked at this window in 20 minutes and doesn't have the code open.
The skill goes beyond review/analysis and instructs the agent to create local files, update TODOS.md, and optionally promote content into repo docs. That materially expands its write authority and creates an unexpected path to alter repository state under the guise of planning, which can be abused for persistence, workflow manipulation, or unauthorized documentation changes.
The instruction to dispatch a reviewer subagent introduces secondary execution and analysis authority not implied by a simple plan-review skill. Subagent orchestration can widen data exposure, make behavior harder to audit, and enable capability escalation if the spawned agent has broader tools or different guardrails.
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
name: plan-ceo-review
description: |
CEO/founder-mode plan review. Rethink the problem, find the 10-star product,
challenge premises, expand scope when it creates a better product. Four modes:
SCOPE EXPANSION (dream big), SELECTIVE EXPANSION (hold scope + cherry-pick
expansions), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials).
Use when asked to "think bigger", "expand scope", "strategy review", "rethink this",
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
name: plan-ceo-review
description: |
CEO/founder-mode plan review. Rethink the problem, find the 10-star product,
challenge premises, expand scope when it creates a better product. Four modes:
SCOPE EXPANSION (dream big), SELECTIVE EXPANSION (hold scope + cherry-pick
expansions), HOLD SCOPE (maximum rigor), SCOPE REDUCTION (strip to essentials).
Use when asked to "think bigger", "expand scope", "strategy review", "rethink this",
The skill directs execution of git/gh and grep commands to inspect repository and PR state, but this operational behavior is not clearly disclosed in the manifest description. Hidden command execution increases surprise and risk, especially where CLI tools may access network resources, private metadata, or credentials.
No suspicious patterns detected.