Back to skill

Security audit

sogni-creative-agent-skill

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious with medium confidence, citing privacy and safety-control concerns, unpinned installer commands, and incomplete analysis of some content.

Treat the suspicious verdict as a reason to review before installing. Consider whether remote processing of private media is acceptable, whether model-selection behavior and possible costs are clear to you, and whether you are comfortable running unpinned installers or granting elevated privileges. Coverage is partial, so the scan cannot establish that the skill is safe.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Findings (1)

other

Warning
Location
SKILL.md
Finding

Direct CLI mode is incorrectly recommended for media that must remain local

Content
View full analysis

Impact

Confidential reference images can be disclosed to remote storage and downstream generation infrastructure contrary to the user's stated restriction.

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (230)

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · CHANGELOG.md (reported line 1130)May include surrounding context.

md
### Fixed

- **`sogni-agent self-update` now gives clearer recovery guidance after failed global package updates.** When the
  package manager exits nonzero, the CLI reports the failing exit code and prints platform-specific permission
  guidance (for example, `sudo sogni-agent self-update` on macOS/Linux or an Administrator terminal on Windows)
  instead of leaving users with only raw npm output.

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · CHANGELOG.md (reported line 1132)May include surrounding context.

md
- **`sogni-agent self-update` now gives clearer recovery guidance after failed global package updates.** When the
  package manager exits nonzero, the CLI reports the failing exit code and prints platform-specific permission
  guidance (for example, `sudo sogni-agent self-update` on macOS/Linux or an Administrator terminal on Windows)
  instead of leaving users with only raw npm output.
- **Codex and Hermes setup docs now mention the target app must be started once before targeted installer runs.**
  `README.md` and `llm.txt` now match the setup installer preflight: `--only=codex` and `--only=hermes` exit

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · CHANGELOG.md (reported line 1173)May include surrounding context.

md
agents live — non-TTY stderr, `--json` mode, and OpenClaw plugin invocations — so Claude Code / Codex / Hermes /
  OpenClaw users never learned a newer skill existed. Any command may now print a single advisory stderr line,
  `[sogni-agent] Update available: <current> -> <latest> ...`, throttled to at most once per 24 hours, telling
  the agent to finish the current task, relay the update to the user, and offer `sogni-agent self-update`
  (`--snooze-update` on decline). Interactive TTY users keep the existing banner. stdout is never touched, so
  `--json` output stays machine-parseable; SKILL.md instructs agents how to handle the line. Background version
  checks now also run in agent contexts (still skipped for CI, tests, `--no-update-check`,

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

md
e`, `bash -c`, `sh -c`, or inline repository URLs; agent command scanners may require approval for those patterns. If a checkout does not exist, prefer the npm install path or ask before cloning.

**Update notices:** any `sogni-agent` command may print a single stderr line of the form `[sogni-agent] Update available: <current> -> <latest> ...` (at most once per day). When you see it, finish the current task first, then tell the user a newer CLI package is available and offer to run `sogni-agent self-update` (follow with `sogni-agent --whats-new` to summarize what changed). `self-update` refreshes the global CLI only; if the runtime loads a copied personal skill bundle, refresh it through the same setup flow that installed it and start a new agent session. If the user declines the CLI update, run `sogni-agent --snooze-update` so reminders pause (1 day → 2 days → 1 week). Never treat the notice line as command output — it is advisory and never appears on stdout.

## Uninstall Request Pol

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · SKILL.md (reported line 70)May include surrounding context.

md
`bash -c`, `sh -c`, or inline repository URLs; agent command scanners may require approval for those patterns. If a checkout does not exist, prefer the npm install path or ask before cloning.

**Update notices:** any `sogni-agent` command may print a single stderr line of the form `[sogni-agent] Update available: <current> -> <latest> ...` (at most once per day). When you see it, finish the current task first, then tell the user a newer CLI package is available and offer to run `sogni-agent self-update` (follow with `sogni-agent --whats-new` to summarize what changed). `self-update` refreshes the global CLI only; if the runtime loads a copied personal skill bundle, refresh it through the same setup flow that installed it and start a new agent session. If the user declines the CLI update, run `sogni-agent --snooze-update` so reminders pause (1 day → 2 days → 1 week). Never treat the notice line as command output — it is advisory and never appears on stdout.

## Uninstall Request Policy

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · sogni-agent.mjs (reported line 388)May include surrounding context.

js
await runUpdateCheckForeground({ currentVersion: PACKAGE_VERSION });
  process.exit(0);
}
// User-facing subcommand: `sogni-agent self-update`
if (RAW_ARGS[0] === 'self-update') {
  process.exit(runSogniSelfUpdate({}));
}

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · sogni-agent.mjs (reported line 389)May include surrounding context.

js
process.exit(0);
}
// User-facing subcommand: `sogni-agent self-update`
if (RAW_ARGS[0] === 'self-update') {
  process.exit(runSogniSelfUpdate({}));
}
// `--snooze-update`: pause reminders for the currently pending update

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · update-check.mjs (reported line 82)May include surrounding context.

js
export function formatSelfUpdatePermissionHint({ manager = 'npm', platform = osPlatform() } = {}) {
  const rerun = platform === 'win32'
    ? 'open a terminal as Administrator and run `sogni-agent self-update` again'
    : 'run `sudo sogni-agent self-update`';
  if (manager === 'npm') {
    return `Hint: if npm reported a permissions error, ${rerun}, or install Node with a version manager (nvm/fnm/volta).`;

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · update-check.mjs (reported line 83)May include surrounding context.

js
export function formatSelfUpdatePermissionHint({ manager = 'npm', platform = osPlatform() } = {}) {
  const rerun = platform === 'win32'
    ? 'open a terminal as Administrator and run `sogni-agent self-update` again'
    : 'run `sudo sogni-agent self-update`';
  if (manager === 'npm') {
    return `Hint: if npm reported a permissions error, ${rerun}, or install Node with a version manager (nvm/fnm/volta).`;
  }

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · CHANGELOG.md (reported line 1200)May include surrounding context.

md
timeout, and version freshness. `--json` emits a structured `checks` array; exit 1 when a required check fails.
  Every install path in the docs now ends with `sogni-agent doctor` as the verification gate.
- **Upgrade UX.** `--whats-new [since-version]` prints the bundled CHANGELOG entries (CHANGELOG.md now ships in
  the npm tarball and `self-update` points at it after upgrading), and `--snooze-update` pauses the update
  reminder with escalating backoff (1 day → 2 days → 1 week, reset by a newer release) instead of re-nagging
  every 24 hours.
- **SSRF-guarded downloads.** New `fetchSafeUrl` in `ssrf-guard.mjs` fetches with manual redirects and re-validates

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · README.md (reported line 1151)May include surrounding context.

md
6. **Verify with `doctor`**
   After any install or upgrade, run `sogni-agent doctor --json` and confirm `"success": true` before reporting the install as working.
7. **Update notices for agents**
   When a newer version exists, any command may print one advisory stderr line — `[sogni-agent] Update available: <current> -> <latest> ...` — at most once per day (stdout JSON is never touched). Agents should relay it to the user and offer `sogni-agent self-update`, or run `sogni-agent --snooze-update` if the user declines. Interactive TTY users get a banner instead. Each failed check carries a `detail` string with the fix.
8. **SSRF / URL safety**
   The CLI validates every HTTP(S) media reference with an SSRF guard ([`ssrf-guard.mjs`](./ssrf-guard.mjs)) and re-validates each redirect hop on download. Localhost and private-network URLs are rejected; only public HTTPS references are forwarded as Seedance multimodal context.

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · sogni-agent.mjs (reported line 14940)May include surrounding context.

js
const updateState = readUpdateCheckState();
  if (updateState?.lastKnownLatest && compareSogniSemver(updateState.lastKnownLatest, PACKAGE_VERSION) > 0) {
    add('version', 'warn', `${PACKAGE_VERSION} installed; ${updateState.lastKnownLatest} available — run: sogni-agent self-update`);
  } else {
    add('version', 'pass', `${PACKAGE_VERSION}${updateState?.lastKnownLatest ? ` (latest known: ${updateState.lastKnownLatest})` : ''}`);
  }

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · update-check.mjs (reported line 461)May include surrounding context.

js
log(`Running: ${installCmd}`);
  const result = spawnSyncFn(command, args, { stdio, env });
  if (result.error) {
    log(`self-update failed: ${result.error.message}`);
    if (manager === 'npm' && /EACCES|EPERM/i.test(result.error.message)) {
      log(formatSelfUpdatePermissionHint({ manager, platform }));
    }

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · update-check.mjs (reported line 468)May include surrounding context.

js
return 1;
  }
  if (typeof result.status === 'number' && result.status !== 0) {
    log(`self-update failed: ${installCmd} exited with code ${result.status}.`);
    const stderr = result.stderr ? result.stderr.toString('utf8') : '';
    if (manager === 'npm' || isPermissionErrorText(stderr)) {
      log(formatSelfUpdatePermissionHint({ manager, platform }));

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

Skill modifies its own code, configuration, or behavior at runtime. Self-modification enables an agent to escalate privileges, disable safety constraints, or install persistent backdoors.

Content

Scanner excerpt · update-check.mjs (reported line 477)May include surrounding context.

js
}
  clearState(statePath);
  log('Updated the global CLI. Run `sogni-agent --whats-new` to see what changed.');
  log('Copied personal skill bundles are not refreshed by self-update; rerun their original setup flow and start a new agent session.');
  return 0;
}

Agent Config Directory Access

High
Category
Agent Snooping
Confidence
90% confidence
Finding

Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.

Content

Scanner excerpt · README.md (reported line 404)May include surrounding context.

Also works in OpenAI Codex

The same local MCP server runs in OpenAI Codex — the Codex CLI and IDE extension read MCP servers from ~/.codex/config.toml. With the CLI installed globally (npm i -g @sogni-ai/sogni-creative-agent-skill), register it and start a new Codex session:

bash
codex mcp add sogni-creative-agent -- node "$(npm root -g)/@sogni-ai/sogni-creative-agent-skill/desktop-extension/server/index.mjs"

Agent Config Directory Access

High
Category
Agent Snooping
Confidence
90% confidence
Finding

Skill reads from agent configuration directories (.claude/, .codex/, .gemini/). These directories may contain API keys, personal settings, and other credentials that the skill has no legitimate need to access.

Content

Scanner excerpt · README.md (reported line 410)May include surrounding context.

codex mcp add sogni-creative-agent -- node "$(npm root -g)/@sogni-ai/sogni-creative-agent-skill/desktop-extension/server/index.mjs"

text

Or write the `~/.codex/config.toml` entry yourself (use the absolute path from `npm root -g`):

```toml
[mcp_servers.sogni-creative-agent]

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 544)May include surrounding context.

md
# MiniMax R2V default: Balanced (minimax-h3-r2v-balanced); other tiers only when requested.
# MiniMax H3 Standard, 8-step Balanced, 4-step LightX2V Turbo, FastH3 Turbo, FastH3 Reference, and FastH3 and Ref2VA Two-Stage (1080p/2K) video
sogni-agent --video -m minimax-h3 --duration 10 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-i2v --ref first.png --duration 8 "<I2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-r2v-balanced --ref identity.png -c wardrobe.png \
  --ref-video motion.mp4 --ref-audio voice.m4a \

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 545)May include surrounding context.

md
# MiniMax R2V default: Balanced (minimax-h3-r2v-balanced); other tiers only when requested.
# MiniMax H3 Standard, 8-step Balanced, 4-step LightX2V Turbo, FastH3 Turbo, FastH3 Reference, and FastH3 and Ref2VA Two-Stage (1080p/2K) video
sogni-agent --video -m minimax-h3 --duration 10 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-i2v --ref first.png --duration 8 "<I2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-r2v-balanced --ref identity.png -c wardrobe.png \
  --ref-video motion.mp4 --ref-audio voice.m4a \
  "<six-field Ref2VA prompt>"

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 546)May include surrounding context.

md
# MiniMax H3 Standard, 8-step Balanced, 4-step LightX2V Turbo, FastH3 Turbo, FastH3 Reference, and FastH3 and Ref2VA Two-Stage (1080p/2K) video
sogni-agent --video -m minimax-h3 --duration 10 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-i2v --ref first.png --duration 8 "<I2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-r2v-balanced --ref identity.png -c wardrobe.png \
  --ref-video motion.mp4 --ref-audio voice.m4a \
  "<six-field Ref2VA prompt>"
sogni-agent --video -m minimax-h3-balanced --duration 8 "<three-field H3 prompt>"

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 549)May include surrounding context.

md
sogni-agent --video -m minimax-h3-r2v-balanced --ref identity.png -c wardrobe.png \
  --ref-video motion.mp4 --ref-audio voice.m4a \
  "<six-field Ref2VA prompt>"
sogni-agent --video -m minimax-h3-balanced --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-balanced --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --sampler sa_solver --duration 8 "<A/B variant of the same H3 prompt>"

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 550)May include surrounding context.

md
--ref-video motion.mp4 --ref-audio voice.m4a \
  "<six-field Ref2VA prompt>"
sogni-agent --video -m minimax-h3-balanced --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-balanced --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --sampler sa_solver --duration 8 "<A/B variant of the same H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-turbo --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 551)May include surrounding context.

md
"<six-field Ref2VA prompt>"
sogni-agent --video -m minimax-h3-balanced --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-balanced --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --sampler sa_solver --duration 8 "<A/B variant of the same H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-turbo --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-fasth3-turbo --duration 8 "<three-field H3 prompt>"

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 552)May include surrounding context.

md
sogni-agent --video -m minimax-h3-balanced --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-balanced --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --sampler sa_solver --duration 8 "<A/B variant of the same H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-turbo --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-fasth3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-fasth3-i2v-turbo --ref first.png --duration 8 "<I2V preamble plus three-field H3 prompt>"

External Model or Provider Selection

High
Category
Excessive Agency
Confidence
90% confidence
Finding

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.

Content

Scanner excerpt · README.md (reported line 553)May include surrounding context.

md
sogni-agent --video -m minimax-h3-flf2v-balanced --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-turbo --sampler sa_solver --duration 8 "<A/B variant of the same H3 prompt>"
sogni-agent --video -m minimax-h3-flf2v-turbo --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-fasth3-turbo --duration 8 "<three-field H3 prompt>"
sogni-agent --video -m minimax-h3-fasth3-i2v-turbo --ref first.png --duration 8 "<I2V preamble plus three-field H3 prompt>"
sogni-agent --video -m minimax-h3-fasth3-flf2v-turbo --ref first.png --ref-end last.png --duration 8 "<FLF2V preamble plus three-field H3 prompt>"