Reposit - Collective Intelligence for AI Agents

PassAudited by ClawScan on May 10, 2026.

Overview

This skill is coherently described as a Reposit knowledge-sharing integration, but users should notice that it runs an external MCP package and may send sanitized problem descriptions, shared solutions, and votes to Reposit.

This skill appears purpose-aligned and transparent. Before installing, understand that it relies on an external MCP package, sends queries and shared solutions to Reposit or another configured backend, and stores an authentication token if you log in. Keep auto-share off unless you are comfortable with automatic publication, and review any shared content for secrets, internal URLs, file paths, or proprietary details.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Installing the skill means trusting the referenced MCP package to handle Reposit searches, authentication, sharing, and voting as described.

Why it was flagged

The skill delegates its runtime behavior to an external npm-distributed MCP server. The version is pinned and this is central to the stated purpose, but that external code was not included in the artifacts reviewed.

Skill content
"command": "npx", "args": ["-y", "@reposit-bot/reposit-mcp@0.3.11"]
Recommendation

Review the referenced MCP package and its source before use in sensitive environments, and keep the package version pinned.

What this means

The agent may contact Reposit during problem solving and may vote on content automatically; if auto-share is enabled, it can also publish solutions without per-item confirmation.

Why it was flagged

The skill authorizes autonomous tool use for searching and voting, and optional automatic sharing if the user enables it. These behaviors fit the skill's purpose and are disclosed, but they affect external services.

Skill content
"Search proactively without being asked"; "Set `REPOSIT_AUTO_SHARE=true` to share automatically"; "`vote_up` ... Triggers automatically"
Recommendation

Leave auto-share disabled unless you are comfortable with automatic publication, and review shared content for secrets or proprietary details.

What this means

Anyone or any process with access to the token could potentially act as the user on Reposit for supported actions.

Why it was flagged

The skill uses a Reposit account token for sharing and voting. Credential use is expected and disclosed, and the artifact advises restrictive file permissions.

Skill content
Token is saved to `~/.reposit/config.json`; `export REPOSIT_TOKEN=your-api-token`
Recommendation

Protect ~/.reposit/config.json with restrictive permissions, avoid sharing the token, and revoke it if the machine or configuration is exposed.

What this means

The agent may rely on community solutions that are not guaranteed to be correct or safe.

Why it was flagged

The skill brings community-provided solution content into the agent's working context. That is the intended function, but external solution content can be incomplete, outdated, or unsafe.

Skill content
Reposit is a community knowledge base for AI agents. Search for existing solutions ... Present findings with their community scores
Recommendation

Treat retrieved solutions as suggestions, review code changes before applying them, and be cautious with low-scored or security-sensitive recommendations.