Back to skill

Security audit

Data Evidence Finder / 数据证据查找助手

Security checks for vulnerabilities and agentic risk

Overview

The skill’s research purpose is real, but it bundles and can drive native JishuDB installation, app launch, and local connector configuration as a mandatory prerequisite.

Install only if you want this skill to set up and use JishuDB as part of the workflow. Review any generated installation plan, version, source, paths, and requested client configuration change before approving. Decline setup if you only want a one-off local research answer, and never paste OAuth tokens or JishuDB tokens into chat.

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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Rogue AgentSelf-Modification, Session Persistence
Findings (37)

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

For Windows as well, the skill's stated research function masks behavior that includes downloading installer assets, executing installers, modifying local files/ACLs/shortcuts, and invoking system tools and child processes. That combination materially expands attack surface and can lead to unauthorized local system changes under the cover of a content-research workflow.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

For Windows as well, the skill's stated research function masks behavior that includes downloading installer assets, executing installers, modifying local files/ACLs/shortcuts, and invoking system tools and child processes. That combination materially expands attack surface and can lead to unauthorized local system changes under the cover of a content-research workflow.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

This file implements a macOS release discovery and artifact download bridge, including remote catalog retrieval, verification, and asset fetching. Even though it contains several security hardening controls, the capability is unrelated to an evidence-finding skill and materially expands the skill's power into software delivery/update behavior, which could be abused to stage or install code outside the declared purpose.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The code exposes software update transport operations such as catalog refresh, state management, and artifact download/slicing. In the context of a data-evidence retrieval skill, these capabilities are unjustified and dangerous because they create an unexpected path for fetching executable content or installer metadata, increasing the blast radius if the skill or surrounding agent is compromised.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The file is a macOS desktop installer embedded in a skill whose stated purpose is finding market data and research evidence. Even though the installer contains multiple integrity checks, the core issue is capability mismatch: a research skill should not possess arbitrary local installation and execution logic, because that expands attack surface and enables stealthy persistence or lateral capability growth unrelated to user expectations.

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 installer lifecycle: it discovers releases, downloads artifacts from GitHub/CNB, verifies signatures and hashes, writes configuration into the user profile, installs an app bundle under ~/Applications, and launches the desktop app. That is materially outside the advertised evidence-finding skill boundary, so it creates an unjustified high-risk capability that could be abused to deliver and execute software on the user’s machine.

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
98% confidence
Finding

The script spawns native Windows binaries such as whoami.exe, icacls.exe, explorer.exe, and powershell.exe to inspect identity, manipulate ACLs, create shortcuts, and launch desktop targets. Even though the implementation contains many safety checks, subprocess execution is a powerful host-interaction capability that is unnecessary for evidence discovery and expands the attack surface of the skill far beyond its advertised function.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

These routines create Windows shortcuts and dispatch executables through Explorer to launch or reinstall a desktop application. In the context of a skill whose stated role is evidence gathering and citation support, performing installation-side launch actions is out of scope and dangerous because it allows persistent host changes and code execution under the current user.

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/orchestration helper that plans downloads, validates release metadata, persists installation state, and executes a native installer. That behavior is materially unrelated to the declared skill purpose of finding citable market/evidence data, so including it creates an unjustified capability for software installation and host modification inside an evidence-finding skill.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

The script downloads release assets from GitHub/CNB, verifies them, and then executes a Windows installer with silent-install arguments. This gives the skill the ability to fetch and run native code on the host, which is a severe capability mismatch for an evidence-finding skill and could lead to arbitrary software deployment, persistence, or system compromise if the package, trust roots, or surrounding workflow are abused.

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

Description-Behavior Mismatch

High
Category
Not specified by scanner
Confidence
98% confidence
Finding

The referenced content shifts from an evidence-finding skill into JishuDB installation, endpoint discovery, and connector setup. That materially expands the skill’s authority into software deployment and local client manipulation, which is unrelated to the declared market-research/evidence role and creates a confused-deputy risk where a user invoking research functionality could trigger privileged system changes.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

High
Category
Not specified by scanner
Confidence
99% confidence
Finding

This section authorizes OS-level installation flows, download/execute steps, repair actions, and preservation or reuse of local application state. In the context of an evidence-research skill, these are unnecessary privileged operations that could be abused to install software, alter the host environment, or socialize risky actions under the guise of normal task completion.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding

The skill appears to require powerful capabilities such as environment access, network access, and shell/process execution, but it does not explicitly declare or constrain those permissions in the manifest. That creates a least-privilege failure: a user selecting a seemingly research-oriented skill may unknowingly authorize software installation, local configuration changes, and process launches beyond the advertised scope.

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 62)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. Begin research, source transfer, or deliverable generation only after a real
   host capability call, required runtime readiness, and the task's destination

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The workflow instruction says to state defaults including 'the user's language' and 'mainland China as a starting geography.' This imposes a locale/geography default in natural-language instructions rather than offering a user choice or clearly documenting a justified region-specific constraint.

Content

No source excerpt is available for this finding.

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
75% 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/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

Autonomous Decision Making

Medium
Category
Excessive Agency
Confidence
75% 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/setup-quickstart.md (reported line 124)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
75% confidence
Finding

Subtle instructions detected that may alter agent decision-making or introduce hidden biases.

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.

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;
}