pinboard-mcp-server
Provides a FastMCP-based Model Context Protocol (MCP) server exposing read-only tools to search and list Pinboard.in bookmarks (including tag listing) for use by LLM clients at inference time. It uses a Pinboard API token from an environment variable and includes caching and rate-limit handling as described in the README.
Score Breakdown
⚙ Agent Friendliness
🔒 Security
Token is provided via environment variable and the README claims tokens are never logged/exposed in error messages. Access is read-only. Scope granularity is effectively all-or-nothing for the token (no fine-grained scopes described). TLS enforcement for the MCP transport and outbound Pinboard calls is not explicitly documented here; score reflects typical secure defaults but uncertainty.
⚡ Reliability
Best When
You have a Pinboard account and want LLM tools to retrieve relevant bookmark metadata quickly and safely (read-only), while respecting Pinboard’s API usage guidance.
Avoid When
You need guaranteed strict recency beyond the described caching strategy, or you cannot securely provide a PINBOARD_TOKEN to the server process.
Use Cases
- • LLM-assisted retrieval of personal knowledge from Pinboard bookmarks
- • Querying bookmarks by search terms (titles/notes/tags)
- • Listing recent bookmarks for context in chat agents
- • Filtering bookmarks by tags and optional date ranges
- • Providing tag statistics to inform retrieval/routing strategies
Not For
- • Writing/modifying Pinboard bookmarks (the service is described as read-only)
- • Use cases requiring real-time freshness on every call without caching
- • Environments where users cannot supply and securely manage a third-party API token
Interface
Authentication
Authentication appears to be a single Pinboard API token supplied via environment variable; no OAuth or scope model is described.
Pricing
Pricing is not described; this appears to be a self-hosted/open-source Python package.
Agent Metadata
Known Gotchas
- ⚠ Rate limits/caching: results freshness may lag due to caching and cache invalidation behavior; agents should not assume real-time updates on every call.
- ⚠ Tool parameters are constrained (e.g., tags list length 1-3; limit max 100); agents should validate/sanitize inputs to avoid tool-argument errors.
Alternatives
Full Evaluation Report
Comprehensive deep-dive: security analysis, reliability audit, agent experience review, cost modeling, competitive positioning, and improvement roadmap for pinboard-mcp-server.
AI-powered analysis · PDF + markdown · Delivered within 30 minutes
Package Brief
Quick verdict, integration guide, cost projections, gotchas with workarounds, and alternatives comparison.
Delivered within 10 minutes
Score Monitoring
Get alerted when this package's AF, security, or reliability scores change significantly. Stay ahead of regressions.
Continuous monitoring
Scores are editorial opinions as of 2026-04-04.