Security audit
Security checks for vulnerabilities and agentic risk
This is a disclosed local transcript-redaction workflow, but users should carefully review any generated transcript before publishing it to GitHub.
Install only if you are comfortable with the agent reading local session logs to prepare a redacted public transcript. Before publishing, review the final rendered transcript for secrets, private paths, unrelated work, and tool output; also verify the installed package actually includes the referenced agent-transcript helper script.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
The workflow includes opening generated preview files via the operating system browser command. If file paths or generated content are not carefully controlled, this can expose users to local command/path abuse or to rendering of untrusted active content, especially because transcript logs are treated as potentially sensitive and may contain attacker-influenced text.
Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
# Codex with explicit model and reasoning
"$AUTOREVIEW" --engine codex --model gpt-5.6-sol --thinking high
# Codex fast mode (priority service tier); needs a model whose catalog lists the tier, silently standard otherwise
"$AUTOREVIEW" --engine codex --codex-speed fast
Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
"$AUTOREVIEW" --engine codex --codex-config 'service_tier="fast"'
# Claude Code aliases or full model names, with optional availability fallback
"$AUTOREVIEW" --engine claude --model claude-fable-5 --thinking max
"$AUTOREVIEW" --engine claude --model claude-fable-5 --fallback-model claude-opus-4-8,claude-sonnet-4-6
# Pi with explicit model and thinking level
Skill selects an external model or provider that may use a different account or billing plan than the operator expects. Undisclosed model switches can cause unexpected cost or quota consumption.
# Claude Code aliases or full model names, with optional availability fallback
"$AUTOREVIEW" --engine claude --model claude-fable-5 --thinking max
"$AUTOREVIEW" --engine claude --model claude-fable-5 --fallback-model claude-opus-4-8,claude-sonnet-4-6
# Pi with explicit model and thinking level
"$AUTOREVIEW" --engine pi --model anthropic/claude-sonnet-4 --thinking high --pi-bin pi
Tool parameters are crafted to achieve unintended or unsafe behavior. Parameter abuse can bypass intended safety checks (e.g. shell=True, --force, dangerous glob patterns).
}
export function deleteUpload(name) {
return execSync(`rm -rf uploads/${name}`);
}
export function publicUser(user) {
Code enumerates, copies, or searches environment variables for secrets. Bulk environment access can collect credentials unrelated to the skill's stated purpose.
def git(repo: Path, *args: str) -> str:
env = os.environ.copy()
env.update(
{
"GIT_AUTHOR_NAME": "Autoreview Test",
Code enumerates, copies, or searches environment variables for secrets. Bulk environment access can collect credentials unrelated to the skill's stated purpose.
def git(repo: Path, *args: str) -> str:
env = os.environ.copy()
env.update(
{
"GIT_AUTHOR_NAME": "Autoreview Test",
Detected: suspicious.dangerous_exec, suspicious.env_credential_access, suspicious.exposed_secret_literal