OpenSkillIndex← back to search
claude-code

Conitens security-audit

Check approval boundaries, data handling, and mutation safety before risky changes are accepted.

install
source · Clone the upstream repo
git clone https://github.com/seunghwaneom/Conitens
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/seunghwaneom/Conitens "$T" && mkdir -p ~/.claude/skills && cp -r "$T/.agents/skills/security-audit" ~/.claude/skills/seunghwaneom-conitens-security-audit && rm -rf "$T"
manifest: .agents/skills/security-audit/SKILL.md
source content

security-audit

Workflow

  1. Identify the write surfaces and approval boundaries involved.
  2. Check for uncontrolled mutation paths, especially around identity and policy.
  3. Flag namespace leaks, unsafe retrieval, or missing approval gates.
  4. Return concrete risks before any sign-off.

Constraints

  • Treat identity edits as high risk.
  • Do not normalize risky actions just because they are convenient.
  • Prefer explicit approval evidence over inferred intent.