| analyze_workspaceA | Scan a local workspace directory and return its project profile. This tool is read-only and performs an offline scan of the target directory to detect
programming languages, active frameworks, dependencies, and project structure signals.
It does not modify any files.
Parameters:
workspace (str): The absolute path to the local directory to be analyzed.
Returns:
dict[str, Any]: A dictionary containing detected languages, frameworks, notable SDKs,
package managers, manifest files, and a high-level summary string of the project profile.
|
| brainstorm_workspaceA | Generate local, file-grounded next-action brainstorm recommendations for a workspace. This tool is read-only and uses local project files, rules, and notes to suggest developer
next steps. It is designed to be run offline.
Parameters:
workspace (str): The absolute path to the local project workspace.
goal (str, optional): A specific engineering or product goal to align recommendations with. Defaults to "".
limit (int, optional): The maximum number of recommendations to return. Defaults to 6.
Returns:
dict[str, Any]: A dictionary containing project summary, suggestions (each with title,
prompt, category, and tag), and execution status flags.
|
| recommend_ecosystemA | Recommend compatible MCP servers, skills, repositories, and APIs for a project. This tool recommends external ecosystem integrations matching the project profile.
It has no side effects.
Parameters:
workspace (str): The absolute path to the local project workspace.
goal (str, optional): The target engineering goal or capability needed. Defaults to "".
limit (int, optional): The maximum number of recommendations to return. Defaults to 9.
live (bool, optional): If True, query public ecosystem indexes for real-time recommendations.
If False (default), uses pre-cached local resources to run entirely offline.
Returns:
dict[str, Any]: A dictionary containing recommended repositories, MCP servers, and hosted API integrations.
|
| build_usecase_blueprintA | Synthesize a 3-layer architecture blueprint (BUILD/INTEGRATE/FOUNDATION) for a goal. This tool is read-only and structures ecosystem findings into an actionable design plan
specifying what should be custom-built vs. integrated.
Parameters:
workspace (str): The absolute path to the local project workspace.
goal (str, optional): The target development or architectural goal. Defaults to "".
limit (int, optional): The maximum number of items to rank and recommend. Defaults to 8.
live (bool, optional): If True, performs live network lookups for ecosystem tools.
If False (default), runs completely offline.
Returns:
dict[str, Any]: A dictionary detailing the 3-layer architectural canvas, confidence scores,
patterns to borrow, and next steps.
|
| scan_research_radarA | Surface academic papers, models, and AI releases relevant to the workspace. This tool is read-only and assists with literature mapping and model selection for agent tasks.
Parameters:
workspace (str): The absolute path to the local project workspace.
goal (str, optional): The research topic or machine learning goal. Defaults to "".
limit (int, optional): The maximum number of papers or models to return. Defaults to 4.
live (bool, optional): If True, queries arXiv and Hugging Face API live.
If False (default), uses local cached results.
Returns:
dict[str, Any]: A dictionary listing relevant papers, machine learning models, and source metadata.
|
| scan_market_radarA | Surface launch directories and developer discussions matching a project goal. This tool is read-only and helps identify user objections, maker positioning, and market feedback.
Parameters:
workspace (str): The absolute path to the local project workspace.
goal (str, optional): The product, feature category, or topic to search. Defaults to "".
limit (int, optional): The maximum number of market signals to return. Defaults to 8.
live (bool, optional): If True, queries Hacker News, DevHunt, and Uneed APIs live.
If False (default), runs completely offline.
Returns:
dict[str, Any]: A dictionary listing developer attention items, categories, and tags.
|
| benchmark_openhumanA | Compare the project's posture and tools against the OpenHuman framework. This tool is read-only and returns a product-level comparison to help define local features.
Parameters:
workspace (str): The absolute path to the local project workspace.
Returns:
dict[str, Any]: A dictionary detailing integration metrics, architectural alignments,
and next-step suggestions.
|
| audit_trust_surfaceA | Audit the project for missing proof links, legacy names, and trust gaps. This tool is read-only and analyzes public-facing metadata and documentation files
(such as README, docs, verify, index.html) to locate credentials and references.
Parameters:
workspace (str): The absolute path to the local project workspace to audit.
target (str, optional): The target trust profile. Must be one of:
- "public-site": A public webpage surface.
- "mcp-server": An MCP server code surface.
- "trust-repo": A public repository layout.
- "repo": A standard repository layout.
- "auto" (default): Auto-detects target based on available files.
Returns:
dict[str, Any]: A dictionary containing a summary of detected proof links, stale naming issues,
gaps in trust coverage, and recommended fixes.
|
| scaffold_editor_integrationA | Preview or write editor configurations to integrate the Matrix Scroll MCP server. This tool is write-capable. By default, it runs in preview-only mode returning a unified diff.
It will only write the configuration to disk if the `write` parameter is explicitly set to True.
Parameters:
workspace (str): The absolute path to the local project workspace.
editor (str): The target editor/host name. Supported values: "cursor", "vscode", "claude".
write (bool, optional): If True, writes the updated configuration directly to the project's
editor settings directory. If False (default), returns a diff preview only.
Returns:
dict[str, Any]: A dictionary containing the target configuration path, config payload,
unified diff preview, and write success status.
|
| plan_matrixscroll_rolloutA | Generate a structured, target-ready rollout plan for the Matrix Scroll protocol. This tool is read-only and outputs a target-specific playbook, objection-handling map,
and compare hooks to make the security story clear to stakeholders.
Parameters:
workspace (str): The absolute path to the local project workspace.
audience (str): The target stakeholder persona. Must be one of:
- "founder": Product-focused value and fast-proof setups.
- "security": Trust limits, sandboxing, and policy enforcement.
- "devrel": Copyable config blocks and developer rollout assets.
- "team": Rollout phases, workspace onboarding, and team config review.
goal (str, optional): The specific engineering or adoption goal. Defaults to "".
Returns:
dict[str, Any]: A dictionary containing target-specific one-liners, steps, proof assets,
common objections with responses, and comparison hooks.
|
