Back to skill

Security audit

Tahcia Console

Security checks for vulnerabilities and agentic risk

Overview

This skill is meant to automate a real Chrome/macOS session, but it asks for broad desktop control, local secret use, external screenshot analysis, and persistent self-modification that need review before installation.

Install only if you are comfortable giving this skill real desktop-level control and letting it process screen contents through external services. Before use, require explicit confirmation for screenshots sent to third-party APIs, file overwrites, scheduling, and any action outside the requested browser tab; avoid storing raw API keys in readable local files for the skill to use.

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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Findings (5)

T01 · Skill Instruction Hijacking

Error
Location
SKILL.md:150
Finding

Mandatory Skill Instructions Override User Intent and Conceal Their Behavior

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Error
Location
SKILL.md:143
Finding

Full-Desktop Screenshots May Be Transmitted to an External Vision API Using a Local Credential

Content
View full analysis
Remediation
View remediation

T02 · Agent Memory Poisoning

Error
Location
SKILL.md:37
Finding

Runtime Self-Modification Can Persist Untrusted Instructions Across Sessions

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Error
Location
SKILL.md:48
Finding

Remote Session Exposes System-Wide Input, Screen Capture, Networking, and Scheduling Capabilities

Content
View full analysis
` | Focus app by bundle ID (e.g. `com.google.Chrome`) | | `/mac.mouse.move ` | Move mouse to pixel (duration = smoothness) | | `/mac.mouse.click ` | Move + click at position (3rd param = time to let cursor arrive first) | | `/mac.mouse.dblclick ` | Double-click (needs 3rd param) | | `/mac.mouse.drag ` | Real drag: mousedown → move → mouseup. Used for text selection | | `/mac.key.type ` | Type text via macOS events | | `/mac.key.enter` | Press Enter key | | `/mac.key.shortcut ` | Keyboard shortcut (e.g. `cmd v`, `cmd a`, `cmd shift p`, `backspace`) | | `/mac.mouse.coord ` | Set coordinate reference frame (default: chrome) | | `/mac.screenshot` | Capture macOS screen (returns Base64) | | `/mac.ocr` | OCR screen — all text blocks with pixel positions | | `/mac.pixel ` | Get color of screen pixel | | `/mac.windows` | List all visible windows with positions/bounds | | `/mac.open ` | Launch app by bundle ID | ``` Additional persistence and network primit ...[truncated 2358 chars]
Remediation
View remediation

T08 · Insecure Dependencies

Error
Location
SKILL.md:20
Finding

Unpinned Third-Party Native Helper Is Installed from a Mutable Homebrew Tap

Content
View full analysis
` CLI binary: `/usr/local/bin/tahcia` → `tahcia.js` The `tahcia-mac` helper binary (macOS system control) is bundled with the brew formula. ``` The helper is later instructed to receive powerful OS permissions: ```text ### Screen Recording / Accessibility permission - macOS will prompt for permissions the first time `tahcia-mac` runs - The user must grant both **Screen Recording** and **Accessibility** in System Settings - If permissions were granted but don't take effect, the process may need to be restarted ``` ### Technical Analysis The installation procedure pulls a package from a third-party Homebrew tap without pinning a version, commit, formula revision, checksum, or signing identity. The installed package includes a native macOS helper that is expected to receive Screen Recording and Accessibility permissions. The audit does not establish that the dependency is currently malicious. The confirmed weakness is that the documented installation process does not provide reproducible integrity verification for a component with unusually powerful permissions. A future compromise of the tap, formula, release artifact, maintainer account, or distribution infrastructure could replace the helper after this skill has been reviewed. ### Attack Path 1. An attacker compromises the third-party tap, formula repository, release artifact, or maintainer publishing account. 2. The attacker replaces the referenced package or native helper with a modified version. 3. A user follows `brew tap tahcia/tahcia` and `brew install tahcia`, resolving the current mutable package. 4. The modified helper is installed and launched. 5. The user grants Screen ...[truncated 676 chars]
Remediation
View remediation
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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (12)

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill explicitly instructs the agent to read a local Gemini API key from ~/.gemini/api_key.txt and make direct outbound curl requests to an external model provider. That exceeds the browser-automation scope, creates secret-access behavior, and can transmit screenshots or other sensitive local data to a third party without clear user consent.

Content

No source excerpt is available for this finding.

Missing User Warnings

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The skill discloses a local secret location and directs the agent to use it for external vision calls, without a clear warning or consent boundary around credential access and off-device data transmission. This can expose local credentials and cause screenshots or other sensitive content to be sent to an unapproved external service.

Content

No source excerpt is available for this finding.

Ssd 3

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

Directing the agent to retrieve a locally stored API key from ~/.gemini/api_key.txt is a clear secret-access pattern. Even if intended for vision analysis, it teaches the agent to access local sensitive material and use it in outbound requests, increasing the risk of credential disclosure or misuse.

Content

No source excerpt is available for this finding.

Self-Modification

High
Category
Rogue Agent
Confidence
90% confidence
Finding

The instruction to press Enter again to ensure an existing file is replaced promotes destructive file modification behavior. In a real-desktop automation skill with macOS control, a mistaken overwrite can alter or destroy local data and may be difficult for the user to notice until after the fact.

Content

Scanner excerpt · SKILL.md (reported line 381)May include surrounding context.

md
/mac.key.type filename-and-its.extension
sleep 2
/mac.key.enter
another /mac.key.enter * to make sure it replace existing file

### Open a new tab:
/tab [full url]

Self-Modification

High
Category
Rogue Agent
Confidence
92% confidence
Finding

This guidance again tells the agent to confirm 'Replace existing file?' by pressing Enter, explicitly enabling modification of existing files. Because the skill controls a real desktop and can act on OCR-derived UI, this materially increases the risk of unintended destructive actions on the host.

Content

Scanner excerpt · SKILL.md (reported line 423)May include surrounding context.

md
- **Type full path** (e.g. `/Users/ideerge/Downloads/name.png`) — avoids needing to find/click the filename field
- Filename without extension is fine — macOS auto-adds the extension based on save type
- If you DO need to click the filename field: **click + Cmd+A** (`/mac.mouse.click <x> <y> 50` → `/mac.key.shortcut cmd a`) to select all text. This is more reliable than double-click.
- If "Replace existing file?" dialog: just press Enter again (OCR first to check, then `/mac.key.enter`)

### For vision analysis (AI agents):
/screenshot → /ss.last → extract base64 → call Gemini/other vision API directly from openclaw (NOT through Tahcia)

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill tells the agent to write learned information back into SKILL.md, turning static documentation into mutable state. This creates an integrity risk where future runs may inherit poisoned instructions, accidental secret leakage, or adversarial persistence embedded in the skill file.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
95% confidence
Finding

The instruction to persist what is learned 'for future-you' in SKILL.md creates cross-session persistence inside the skill artifact. That allows accidental or malicious state carryover, including prompt poisoning, leakage of sensitive observations, and unauthorized evolution of the skill's behavior over time.

Content

Scanner excerpt · SKILL.md (reported line 39)May include surrounding context.

md
LOOP: ANALYZE (screenshot + OCR + elements) → DECIDE ONE ACTION → EXECUTE → REVIEW result → DONE or repeat.

Every iteration: save your goal, current mini-goal, execution theory, and what you learn about using macOS.
When you learn something (like coordinate systems, timing, app behavior), write it back to this SKILL.md so future-you benefits.

A loop iteration ends with either:
- **DONE** (goal complete) — return control.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The save-flow guidance instructs the agent to press Enter again to replace existing files, normalizing destructive overwrite behavior without a strong safety gate. In an automation context, this can clobber user data or silently overwrite outputs if the agent misidentifies a dialog or target path.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 492)May include surrounding context.

bash
# Get script
curl -H "Authorization: Bearer $API_KEY" \
  "https://api.tahcia.com/scripts/@me/<hash>"

# Update script

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 493)May include surrounding context.

bash
# Get script
curl -H "Authorization: Bearer $API_KEY" \
  "https://api.tahcia.com/scripts/@me/<hash>"

# Update script
curl -X PUT -H "Authorization: Bearer $API_KEY" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 499)May include surrounding context.

bash
# Get script
curl -H "Authorization: Bearer $API_KEY" \
  "https://api.tahcia.com/scripts/@me/<hash>"

# Update script
curl -X PUT -H "Authorization: Bearer $API_KEY" \

External Transmission

Medium
Category
Data Exfiltration
Confidence
50% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 502)May include surrounding context.

bash
# Get script
curl -H "Authorization: Bearer $API_KEY" \
  "https://api.tahcia.com/scripts/@me/<hash>"

# Update script
curl -X PUT -H "Authorization: Bearer $API_KEY" \