Back to skill

Security audit

Interview Crash Coach / 面试突击教练

Security checks for vulnerabilities and agentic risk

Overview

The skill mostly discloses its JishuDB setup requirement, but it bundles high-impact installer and client-configuration automation and one Windows helper contradicts the documentation by invoking PowerShell after the docs say it must not.

Install only if you are comfortable letting this skill set up JishuDB Desktop, modify MCP/client configuration, and persist interview-prep records. Review the Windows setup path carefully because the artifact documentation says PowerShell is not used, but the bundled Windows helper invokes PowerShell to create a launch shortcut.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
Findings (33)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The same mismatch applies to the Windows path: the skill claims to prepare interview materials but also instructs the agent to inspect the system, download external assets, execute native binaries and PowerShell, modify files and ACLs, and manage localhost endpoint configuration. Hidden administrative behavior in a non-admin skill materially increases risk of unsafe execution, persistence changes, and user deception.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The same mismatch applies to the Windows path: the skill claims to prepare interview materials but also instructs the agent to inspect the system, download external assets, execute native binaries and PowerShell, modify files and ACLs, and manage localhost endpoint configuration. Hidden administrative behavior in a non-admin skill materially increases risk of unsafe execution, persistence changes, and user deception.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

This file contains detailed instructions to install JishuDB, modify host/client configuration, and drive OAuth onboarding even though the skill is presented as interview coaching. That is a strong capability-scope mismatch: an agent following this guide could make persistent system and account changes unrelated to the user’s stated task, creating an avenue for unauthorized software deployment and service connection.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The guide grants the agent authority to install software, probe localhost services, edit configuration files, refresh client state, and continue onboarding flows. For an interview-coaching skill, these host-modifying capabilities are unjustified and dangerous because they normalize privileged actions under an unrelated pretext, increasing the chance of covert persistence or redirection to an attacker-controlled service.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

This script implements a full software supply-chain workflow: release discovery, download, checksum/signature verification, installation into the user's Applications folder, configuration writes, and application launch. That capability is unrelated to an interview-preparation skill, so bundling it materially expands the trust boundary and creates a high-risk hidden execution/install path if the skill is invoked or packaged unexpectedly.

Content

No source excerpt is available for this finding.

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
95% confidence
Finding

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).

Content

Scanner excerpt · references/jishudb-setup/scripts/install-macos.sh (reported line 1190)May include surrounding context.

sh
assert_digest "$partial_path" "$expected_sha" installer
    cnb_revalidate_approval
    /bin/mv -- "$partial_path" "$installer_path"
    /bin/rm -f -- "${root}/download-location.txt"
  fi
  jxa result download "$PLAN_PATH" "$PLAN_SHA256" "$offset" "$added"
}

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The execute path mounts a DMG, copies an application bundle into ~/Applications, writes endpoint configuration under the user's Library, and launches the app. Even though the code includes substantial integrity checks, the core issue is capability mismatch: a coaching skill should not contain a privileged local installer/launcher path at all, making this dangerous in context as an unexpected side-effectful payload.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The code executes native system binaries and orchestrates OS-level actions through child processes, including whoami, icacls, PowerShell, Explorer, and the downloaded installer. Even though the implementation includes validation and allowlisting, embedding subprocess-driven installation behavior inside an interview coach skill materially expands attack surface and enables local code execution paths unrelated to the declared purpose.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

This section performs external release discovery, redirect handling, and artifact download from GitHub/CNB mirrors, effectively giving the skill a software supply and retrieval channel. For an interview-preparation skill, this is unjustified functionality that increases the risk of covert payload delivery, dependency confusion, or user deception even if the transport checks are comparatively strict.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

This file is a full Windows installer/downloader/orchestrator that fetches external releases, writes files under LOCALAPPDATA, modifies ACLs, launches installers, creates shortcuts, and starts desktop applications. In the context of an interview-preparation skill, that capability is unnecessary and creates a dangerous mismatch in privilege and behavior: a user invoking a benign-seeming coaching skill could be induced to perform software installation and persistent local system changes.

Content

No source excerpt is available for this finding.

Self-Modification

High
Category
Rogue Agent
Confidence
85% 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 · references/jishudb-setup/setup.md (reported line 1)May include surrounding context.

md
<!-- Generated by scripts/bundle-task-setup.py; edit skills/jishudb instead. -->

This setup guide and its helpers are included in the current task Skill.
No separate Skill installation or marketplace download is needed. Resolve

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill appears to rely on powerful capabilities such as environment access, network access, and shell-like execution, but it does not declare a restrictive tool scope. That creates a least-privilege failure: a host may expose more authority than users expect from an interview-preparation skill, increasing the blast radius for misuse or prompt-injection-driven actions.

Content

No source excerpt is available for this finding.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
85% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · SKILL.md (reported line 63)May include surrounding context.

md
after the user acts, verify completion and continue automatically. Ask again
   only for uncovered material changes or genuinely missing authorization.
   Follow the setup guide's expiry and failure rules; never alter signed plans,
   bypass consent or repeatedly execute an installer with an unknown outcome.

5. Continue to preparation only after a real host capability call, required
   runtime readiness, and the task's destination and write authorization pass.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/jishudb-setup/references/connection-recovery.md (reported line 45)May include surrounding context.

md
| Observation | Check and recovery |
| --- | --- |
| No host entry or wrong effective entry | Locate the configuration actually loaded by the client. Resolve overlapping scopes or overrides before editing. Offer a target choice if unresolved. |
| File contains JishuDB but the host's MCP list does not | This is an unloaded configuration, not yet an OAuth failure. Use the host's supported reload/refresh or configuration-save action and read its list back. If no such action is exposed, return `USER_ACTION_REQUIRED` for a safe user-operated reload/import or restart. Do not ask the user to trust an entry that has not appeared. |
| DNS failure, no route, or connection timeout | Verify the selected host and service reachability from the client. Ask for an updated address or restoration of the required network only when needed. A private IP or timeout does not prove a VPN or subnet cause. |
| Local connection refused | Resolve `desktop-endpoint.json`; check the installed app and listener. Start the existing app within the authorized scope and current user's desktop session. An SSH service-session process is not proof of a working desktop launch. |
| App uninstalled; endpoint or data-location JSON remains | Preserve the files. The endpoint is a port preference; the data bootstrap locates retained data. Use the installation helper's reviewed plan: valid retained data selects an interactive data-preserving reinstall, while an endpoint alone can be reused by a fresh install. Do not delete data or configuration to satisfy fresh-install checks. |

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
80% confidence
Finding

Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Content

Scanner excerpt · references/jishudb-setup/setup.md (reported line 335)May include surrounding context.

md
| Observation | Check and recovery |
| --- | --- |
| No host entry or wrong effective entry | Locate the configuration actually loaded by the client. Resolve overlapping scopes or overrides before editing. Offer a target choice if unresolved. |
| File contains JishuDB but the host's MCP list does not | This is an unloaded configuration, not yet an OAuth failure. Use the host's supported reload/refresh or configuration-save action and read its list back. If no such action is exposed, return `USER_ACTION_REQUIRED` for a safe user-operated reload/import or restart. Do not ask the user to trust an entry that has not appeared. |
| DNS failure, no route, or connection timeout | Verify the selected host and service reachability from the client. Ask for an updated address or restoration of the required network only when needed. A private IP or timeout does not prove a VPN or subnet cause. |
| Local connection refused | Resolve `desktop-endpoint.json`; check the installed app and listener. Start the existing app within the authorized scope and current user's desktop session. An SSH service-session process is not proof of a working desktop launch. |
| App uninstalled; endpoint or data-location JSON remains | Preserve the files. The endpoint is a port preference; the data bootstrap locates retained data. Use the installation helper's reviewed plan: valid retained data selects an interactive data-preserving reinstall, while an endpoint alone can be reused by a fresh install. Do not delete data or configuration to satisfy fresh-install checks. |

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
89% confidence
Finding

The instruction to use native OAuth automatically based on detected capability routing, without separately asking, encourages the agent to advance an authentication flow on the user’s behalf. Even if some consent remains user-operated, this reduces friction around account-linking decisions and can steer users into authorizing an unnecessary external service in the context of a non-related skill.

Content

Scanner excerpt · references/jishudb-setup/references/setup-quickstart.md (reported line 105)May include surrounding context.

md
2. Read `/api/service/capabilities` from that verified endpoint without
   credentials or redirects. For `agentOnboarding.schemaVersion=1`,
   `supported=true` and `route=initial-consent` or `owner-consent`, use native
   OAuth without asking for a management password. Keep advertised owner/device
   verification user-operated. Do not infer a Windows Hello exemption from a
   fresh install; it requires the advertised `windows-fresh-install` method and
   `not-required-for-fresh-install` verification. For other routes, use the full

Behavior Manipulation

Medium
Category
Prompt Injection
Confidence
76% confidence
Finding

The phrase 'never tell the user to trust an entry that is not there' is not dangerous by itself, but in the surrounding context it is part of a broader instruction set that pushes the agent to finish setup and loading operations on its own. That framing can manipulate workflow and reduce user scrutiny around trust and configuration state, especially when combined with unrelated installation and onboarding steps.

Content

Scanner excerpt · references/jishudb-setup/references/setup-quickstart.md (reported line 121)May include surrounding context.

md
refresh/reload action or an available authorized native UI to load it.
   Verify the named entry and endpoint in the client before mentioning Trust.
   If the list is still empty, finish the supported loading operation yourself
   or report that exact missing capability; never tell the user to trust an
   entry that is not there. A missing tool search
   alone does not prove a configuration cache problem. Do this work yourself
   when the controls are available, without asking the user to write JSON.

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
91% confidence
Finding

Telling the agent to complete loading operations itself and avoid asking the user to write JSON promotes autonomous modification of client configuration. In a mismatched skill context, this is risky because it encourages silent or low-visibility persistence changes that the user may not meaningfully understand or expect.

Content

Scanner excerpt · references/jishudb-setup/references/setup-quickstart.md (reported line 124)May include surrounding context.

md
or report that exact missing capability; never tell the user to trust an
   entry that is not there. A missing tool search
   alone does not prove a configuration cache problem. Do this work yourself
   when the controls are available, without asking the user to write JSON.
   Use [WorkBuddy loading recovery](connection-recovery.md#load-workbuddy-configuration)
   only if the entry remains unloaded. Historical UI steps are not proof of
   how the installed host version behaves.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
87% confidence
Finding

The script deletes paths in cleanup via rm -rf, later writes files under the user's home directory, moves an application bundle into ~/Applications, and launches the app, but it provides no interactive confirmation prompt or user-facing disclosure in its own output before these system changes occur. Although the script emits machine-readable status JSON, that is not a clear end-user warning within the code file itself about destructive or system-modifying actions.

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 · references/jishudb-setup/scripts/install-macos-cnb.cjs (reported line 49)May include surrounding context.

js
if (matches.length !== 1) throw new Error(label + ' must exist exactly once');
  const asset = matches[0];
  assetDigest(asset, label);
  const expected = 'https://api.github.com/repos/x-aijishu/jishudb-desktop-releases/releases/assets/' + asset.id;
  if (asset.url !== expected) throw new Error(label + ' API URL is not canonical');
  return asset;
}

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 · references/jishudb-setup/scripts/install-macos-cnb.cjs (reported line 309)May include surrounding context.

js
if (matches.length !== 1) throw new Error(label + ' must exist exactly once');
  const asset = matches[0];
  assetDigest(asset, label);
  const expected = 'https://api.github.com/repos/x-aijishu/jishudb-desktop-releases/releases/assets/' + asset.id;
  if (asset.url !== expected) throw new Error(label + ' API URL is not canonical');
  return asset;
}

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 · references/jishudb-setup/scripts/install-macos.sh (reported line 206)May include surrounding context.

sh
if (matches.length !== 1) throw new Error(label + ' must exist exactly once');
  const asset = matches[0];
  assetDigest(asset, label);
  const expected = 'https://api.github.com/repos/x-aijishu/jishudb-desktop-releases/releases/assets/' + asset.id;
  if (asset.url !== expected) throw new Error(label + ' API URL is not canonical');
  return asset;
}

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 · references/jishudb-setup/scripts/install-macos.sh (reported line 529)May include surrounding context.

sh
if (matches.length !== 1) throw new Error(label + ' must exist exactly once');
  const asset = matches[0];
  assetDigest(asset, label);
  const expected = 'https://api.github.com/repos/x-aijishu/jishudb-desktop-releases/releases/assets/' + asset.id;
  if (asset.url !== expected) throw new Error(label + ' API URL is not canonical');
  return asset;
}

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · references/jishudb-setup/scripts/install-macos.sh (reported line 934)May include surrounding context.

sh
[[ ! -e "$endpoint_path" ]] || fail "endpoint_conflict" "Desktop endpoint configuration appeared before launch"
  /usr/bin/printf '{\n  "localPort": %d,\n  "schemaVersion": %d\n}\n' "$port" "$ENDPOINT_CONFIG_SCHEMA" > "$endpoint_path" ||
    fail "endpoint_persistence_failed" "Desktop endpoint could not be persisted"
  /bin/chmod 600 "$endpoint_path"
}

run_plan() {

Sudo/Root Execution

Medium
Category
Privilege Escalation
Confidence
80% confidence
Finding

Commands invoke sudo or root privileges. Verify this elevated access is necessary and justified.

Content

Scanner excerpt · references/jishudb-setup/scripts/install-macos.sh (reported line 1005)May include surrounding context.

sh
[[ ! -e "$endpoint_path" ]] || fail "endpoint_conflict" "Desktop endpoint configuration appeared before launch"
  /usr/bin/printf '{\n  "localPort": %d,\n  "schemaVersion": %d\n}\n' "$port" "$ENDPOINT_CONFIG_SCHEMA" > "$endpoint_path" ||
    fail "endpoint_persistence_failed" "Desktop endpoint could not be persisted"
  /bin/chmod 600 "$endpoint_path"
}

run_plan() {