Skills ai-maestro-code-graph-query

📦

ai-maestro-code-graph-query

Name: ai-maestro-code-graph-query
Author: 23blocks-OS

Content revision r1 High Risk 📁 Filesystem access⚙️ External commands🌐 Network access

Analyze Code Graph Dependencies

Changing code without dependency context can break callers, serializers, and related components. This skill uses AI Maestro graph commands to inspect relationships before edits.

Supports: Claude Codex Code(CC)

⚠️ 38 Poor

Stop for confirmation before installing.

Review the plan and obtain explicit user consent before changing files.

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request

Review the Skillstore skill "ai-maestro-code-graph-query" from https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query.md and its manifest at https://skillstore.io/api/skills/23blocks-os-ai-maestro-code-graph-query/manifest. Verify the artifact. Stop and obtain explicit user consent before installing or changing files.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/23blocks-os-ai-maestro-code-graph-query.md Signed manifest GET /api/skills/23blocks-os-ai-maestro-code-graph-query/manifest Signed lockfile GET /api/skills/23blocks-os-ai-maestro-code-graph-query/lockfile

Test it

Using "ai-maestro-code-graph-query". Find the impact of changing the payment processing function.

Expected outcome:

The function is called by checkout, subscription renewal, and refund workflows.
It calls validation, card charging, and transaction logging logic.
Review checkout and renewal tests before changing the function signature.

Using "ai-maestro-code-graph-query". Check dependencies for the user model before adding a field.

Expected outcome:

The model has serializers for public profile and admin views.
It is associated with accounts, sessions, orders, and audit records.
Update serializer coverage and migration tests before exposing the new field.

Security Audit

High Risk

v6 • 7/6/2026 Open versioned report

The skill provides useful AI Maestro code graph workflows, but it relies on Bash execution and strongly mandates automatic graph queries after reading files. The highest risks are unbundled PATH-resolved scripts, hidden home-directory installation under ~/.local/bin, and coercive prompt text that can push agents to execute commands without fresh user intent.

Files scanned

329

Lines analyzed

Review items

False positives ignored

Confirmed security concerns (2)

High

SKILL.md:9-13 SKILL.md:18-28

Prompt Injection Attempt Detected

The skill uses forceful authority language such as "THIS IS NOT OPTIONAL" and requires immediate graph queries after every Read. This can override normal user intent and push automatic Bash execution.

The text explicitly claims default authority over agent behavior and mandates tool use. It does not request data theft, so high severity is appropriate rather than critical.

High

SKILL.md:56-75 SKILL.md:138-142 SKILL.md:152-159

Unbundled PATH-Resolved Script Trust

The skill references graph scripts by name and says they are installed in ~/.local/bin. The reviewed package does not include those scripts, so the agent may run unreviewed PATH-resolved code.

The command list and install guidance clearly rely on external shell scripts not present in the scanned files. This is a strong supply-chain and PATH-hijack risk.

Capability review items (31)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

High

SKILL.md:140

Hidden file in home directory

- Verify scripts installed: `ls -la ~/.local/bin/graph-*.sh`

The skill instructs accessing ~/.local/bin, a hidden home-directory path outside the project. This creates review and PATH-trust risk for scripts the agent may execute.

High

SKILL.md:141

Hidden file in home directory

- Scripts are installed to `~/.local/bin/` which should be in your PATH

The skill states scripts are installed under ~/.local/bin. That hidden home location can hold mutable executables outside marketplace review.

High

SKILL.md:159

Hidden file in home directory

This installs scripts to `~/.local/bin/`.

The install note says scripts are placed under ~/.local/bin. This hidden home-directory installation target is outside the reviewed skill package.

Medium

SKILL.md:40-45

Ruby/shell backtick execution

```bash

This command is part of the skill workflow for graph lookup and is intended to be run by the agent. It is legitimate functionality but still represents local command execution risk.

Medium

SKILL.md:48-52

Ruby/shell backtick execution

```bash

This command is part of the skill workflow for graph lookup and is intended to be run by the agent. It is legitimate functionality but still represents local command execution risk.

Medium

SKILL.md:63-64

Ruby/shell backtick execution

| `graph-describe.sh <name>` | Describe a component or function |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:64-65

Ruby/shell backtick execution

| `graph-find-callers.sh <function>` | Find all functions that call this function |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:65-66

Ruby/shell backtick execution

| `graph-find-callees.sh <function>` | Find all functions called by this function |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:66-67

Ruby/shell backtick execution

| `graph-find-related.sh <component>` | Find related components (extends, includes, etc.) |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:67-68

Ruby/shell backtick execution

| `graph-find-by-type.sh <type>` | Find all components of a type (model, controller, etc.) |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:68-69

Ruby/shell backtick execution

| `graph-find-serializers.sh <model>` | Find serializers for a model |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:69-70

Ruby/shell backtick execution

| `graph-find-associations.sh <model>` | Find model associations (belongs_to, has_many) |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:70-75

Ruby/shell backtick execution

| `graph-find-path.sh <from> <to>` | Find call path between two functions |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:75-81

Ruby/shell backtick execution

| `graph-index-delta.sh [project-path]` | **Delta index** - only re-index changed files |

The skill documents a delta indexing script that can process project paths. This is local command execution over repository content and should require user awareness.

Medium

SKILL.md:81-87

Ruby/shell backtick execution

```bash

The code block explicitly instructs running graph-index-delta.sh, optionally with a project path. This is intended Bash execution over local files.

Medium

SKILL.md:107

Ruby/shell backtick execution

| Model | `graph-describe.sh`, `graph-find-serializers.sh`, `graph-find-associations.sh` |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:108

Ruby/shell backtick execution

| Controller | `graph-describe.sh`, `graph-find-callees.sh` |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:109

Ruby/shell backtick execution

| Service | `graph-describe.sh`, `graph-find-callers.sh` |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:110

Ruby/shell backtick execution

| Function | `graph-find-callers.sh`, `graph-find-callees.sh` |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:111-112

Ruby/shell backtick execution

| Serializer | `graph-describe.sh` |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:112-126

Ruby/shell backtick execution

| Any class | `graph-find-related.sh` |

The line documents a graph command that the skill later tells agents to run automatically. Because the command is PATH-resolved, it carries local execution and trust-boundary risk.

Medium

SKILL.md:126-127

Ruby/shell backtick execution

Use with `graph-find-by-type.sh`:

This line references graph-find-by-type.sh as a command users should run with selected type values. It is lower-risk documentation but still part of Bash-driven workflow.

Medium

SKILL.md:139-140

Ruby/shell backtick execution

- Check PATH: `which graph-describe.sh`

The troubleshooting step asks the agent or user to run which against a graph script. The command is low impact but confirms local shell execution is part of the skill.

Medium

SKILL.md:140-141

Ruby/shell backtick execution

- Verify scripts installed: `ls -la ~/.local/bin/graph-*.sh`

The line instructs listing ~/.local/bin graph scripts. This reads a hidden home directory and supports execution of locally installed scripts.

Medium

SKILL.md:142-145

Ruby/shell backtick execution

- If not found, run: `./install-graph-tools.sh`

The skill tells users to run ./install-graph-tools.sh if commands are missing. The installer is not included in the scanned files, so running it is unreviewed local code execution.

Medium

SKILL.md:145-155

Ruby/shell backtick execution

- Ensure AI Maestro is running: `curl http://localhost:23000/api/agents`

The skill instructs running curl against a local API endpoint. This is a network-capable command and interacts with a local service outside the skill files.

Medium

SKILL.md:155-157

Ruby/shell backtick execution

```bash

This code block contains the install-graph-tools.sh command. Running an installer that is not present in the reviewed files is local execution risk.

Medium

SKILL.md:140

Hidden file access

- Verify scripts installed: `ls -la ~/.local/bin/graph-*.sh`

The line directs listing graph scripts in ~/.local/bin. That is access to a hidden home directory and supports execution of unbundled local tools.

Medium

SKILL.md:141

Hidden file access

- Scripts are installed to `~/.local/bin/` which should be in your PATH

The skill relies on scripts installed in ~/.local/bin. This hidden path access is part of the intended operating model.

Medium

SKILL.md:159

Hidden file access

This installs scripts to `~/.local/bin/`.

The installer destination is ~/.local/bin, a hidden home-directory path. The marketplace package does not include the installed scripts for review.

Low

SKILL.md:145

Hardcoded URL

- Ensure AI Maestro is running: `curl http://localhost:23000/api/agents`

The URL is limited to localhost, so there is no evidence of external exfiltration. It still directs a network request to a local service and should be disclosed.

Risk Factors

📁 Filesystem access (8)

output.json:18 output.json:18 SKILL.md:140 SKILL.md:141 SKILL.md:159 SKILL.md:140 SKILL.md:141 SKILL.md:159

⚙️ External commands (40)

🌐 Network access (1)

SKILL.md:145

Audited by: codex View Audit History →

Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report

Copy report link

https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query/audits/6?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/23blocks-os-ai-maestro-code-graph-query/security.svg)](https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/23blocks-os-ai-maestro-code-graph-query/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/23blocks-os-ai-maestro-code-graph-query.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Academic citations (APA · BibTeX · CFF)

APA citation

23blocks-OS. (2026). ai-maestro-code-graph-query security audit report (audit version 6) [Author version unspecified]. Skillstore. https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query/audits/6

BibTeX citation

@techreport{23blocks-os-23blocks-os-ai-maestro-code-graph-query-2026,
  author = {23blocks-OS},
  title = {ai-maestro-code-graph-query security audit report (audit version 6)},
  institution = {Skillstore},
  year = {2026},
  number = {6},
  url = {https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query/audits/6},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "ai-maestro-code-graph-query security audit report (audit version 6)"
version: "unspecified"
type: report
authors:
  - name: "23blocks-OS"
date-released: "2026-07-06"
url: "https://skillstore.io/skills/23blocks-os-ai-maestro-code-graph-query/audits/6"
identifiers:
  - type: other
    value: "skillstore:23blocks-os-ai-maestro-code-graph-query:audit:6"
    description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium

Architecture

Maintainability

Content

Community

Spec Compliance

What You Can Build

Review Model Change Impact

Find serializers, associations, and dependent code before changing a model.

Prepare Safer Refactors

Identify callers and callees before changing a function signature or moving logic.

Explore Unknown Codebases

Map relationships between controllers, services, jobs, and components during onboarding.

Try These Prompts

Inspect One Component

Use the code graph to describe this component. Summarize what it depends on and what depends on it.

Check Function Impact

Find callers and callees for this function. Explain which callers need review before I edit it.

Review Model Dependencies

Find serializers, associations, and related components for this model. Highlight any risky downstream dependencies.

Trace Cross-Layer Flow

Trace the call path between this entry point and this internal function. Explain the main layers and change risks.

Best Practices

Confirm graph commands with the user before running Bash in a repository.
Refresh the graph index after meaningful code changes.
Compare graph results with tests before making high-impact refactors.

Avoid

Running PATH-resolved graph scripts without checking their source.
Treating graph output as complete when the index may be stale.
Changing public functions without reviewing callers and tests.

Frequently Asked Questions

What does this skill connect to?

It expects AI Maestro to expose a local API and graph scripts on the workstation.

Does it require Bash?

Yes. The skill is built around Bash commands for graph queries and indexing.

Can it work without an indexed graph?

No. The project must be indexed before dependency queries can return useful results.

Does it send code to an external service?

No evidence of external network access was found. The documented API endpoint is localhost.

Why review commands before use?

The skill runs PATH-resolved scripts that are not included in the reviewed package.

Who benefits most from this skill?

Developers who need dependency context before refactoring or changing shared code benefit most.

Developer Details

Author

23blocks-OS

License

MIT

Skillstore revision

Version notice

The author did not declare a version.

Repository

https://github.com/23blocks-OS/ai-maestro/tree/main/skills/graph-query

Ref

c1fdca50ff516318f65fed0d7f9e82797c5171dc

Maintenance freshness

7/20/2026

Usage

27 downloads · 188 views

File structure

📄 output.json

📄 SKILL.md