saagarpatel.dev Portfolio
Server Details
Agent-native MCP server over the public saagarpatel.dev corpus. Read-only, stateless.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-11-25
- URL
- Repository
- saagpatel/portfolio-mcp
- GitHub Stars
- 0
- Server Listing
- portfolio-mcp
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 12 tools
The get_/list_ tools are cleanly separated by resource+action, but three pairs are near-duplicates: get_operant_results vs show_operant_results, and list_projects vs show_project_gallery vs show_project_timeline all surface the same underlying project data. Descriptions explain that the show_ variants are MCP Apps renderings that fall back to the same JSON, which helps, but an agent cannot know whether the host supports MCP Apps, so the choice remains genuinely ambiguous.
All names are snake_case verb_noun and follow a readable convention: get_* for single-resource reads, list_* for collections, show_* for rendered views, plus search and play_demo. The only wrinkle is that get_/list_/show_ encode presentation rather than action, so the verb prefix set is slightly irregular, but nothing is chaotic.
Twelve tools sits comfortably in the well-scoped 3-15 band for a content-plus-portfolio server. The count is inflated by four rendering duplicates (show_operant_results, show_project_gallery, show_project_timeline), so it is slightly redundant rather than bloated.
Coverage is strong for a read-only site surface: discovery (list_corpus, search), retrieval (get_document, get_profile, get_repo_profile), projects/repos listing, plus demos and benchmark data. Minor gaps: there is no get_project by id to complement list_projects, and no list_demos to enumerate playable demos for discovery.
Available Tools
12 toolsget_documentGet a documentARead-onlyInspect
Return the full Markdown of one document by id, e.g. "writing/the-handoff" or "book/preface". Get ids from search or list_corpus.
Inputs: id (required).
get_operant_resultsGet OPERANT benchmark resultsARead-onlyInspect
Return the public, sanitized OPERANT results: per-model operator-calibration-score (OCS) profiles, the headline figures, and the calibration-profiles (not a flat leaderboard) framing. Sourced from the published OPERANT mirror. Returns availability=false if no dataset was baked.
Inputs: no named parameters.
get_profileGet profileARead-onlyInspect
Return Saagar Patel's profile: name, one-line summary, site, and the about / now / uses pages. Start here to understand who this is and what they work on.
Inputs: no named parameters.
get_repo_profileGet repo profileARead-onlyInspect
Return one public-safe repo answering profile by repo_id, including purpose, safe commands, agent surfaces, proof refs, freshness, exclusions, limitations, and integrity metadata.
Inputs: repoId (required).
list_corpusList the corpusARead-onlyInspect
List every document with its id, title, type, date, and word count, plus per-type counts. The table of contents for the whole site. Optionally filter by type.
Inputs: type.
list_projectsList projectsBRead-onlyInspect
List the curated, public projects with stack, status, last-active date, and test/CI signals, plus anonymized portfolio-wide aggregates. Optionally filter by status.
Inputs: limit, status.
list_repo_profilesList repo profilesARead-onlyInspect
List public-safe repo answering profiles with repo id, attention state, freshness, and proof counts. Profiles are static artifacts baked from saagarpatel.dev, not live repo health.
Inputs: limit, attentionState.
play_demoPlay a portfolio demoARead-onlyInspect
Play one of the site's interactive demos, rebuilt from the public corpus pages that fully describe them. "interleaving": the Interleaving Explorer, two writers on one database row, six schedules, warn vs enforce mode, and the lost update with its receipts. "failure-museum": match each Agent Failure Museum exhibit (synthetic failures) to the control that catches it. Hosts that support MCP Apps render it as an interactive view; others get the same data as JSON.
Inputs: demo.
searchSearch the writing corpusARead-onlyInspect
Full-text (BM25) search across essays, book chapters, field notes, and pages. Returns ranked matches with a snippet, id, and canonical URL. Pass an id to get_document to read the full text.
Inputs: query (required), limit, section.
show_operant_resultsShow OPERANT resultsBRead-onlyInspect
Show the public OPERANT benchmark as calibration-profile cards (historical OCS repeats, cross-provider TPR/FPR profiles, and the metric) with a playable demo: the reader acts as the operator's agent on six cases and is scored with the real OCS = TPR - FPR. The rows are historical calculations, not a leaderboard. Hosts that support MCP Apps render it as an interactive view; others get the same data as JSON.
Inputs: section.
show_project_galleryShow project galleryARead-onlyInspect
Show the curated public projects as a gallery of cards: name, category, stack, lifecycle, last-active date, test/CI signals, and the public repository link, with archive-wide totals. Hosts that support MCP Apps render it as an interactive gallery; others get the same data as JSON.
Inputs: status, lifecycle.
show_project_timelineShow project timelineARead-onlyInspect
Show the curated public projects on a timeline grouped by last-active month, colored by lifecycle, with the lifecycle breakdown of the whole archive. Hosts that support MCP Apps render it as an interactive timeline; others get the same data as JSON.
Inputs: no named parameters.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Read-only MCP server exposing a user ORANO library to their own AI agent.
Read-only, no-auth MCP server over Human x AI's Content Hub and confirmed 2027 facts.
Agent-native MCP server over 49M+ US public and government records, privacy-first, always current.
Related MCP Servers
- AlicenseAqualityBmaintenanceA read-only MCP server so AI agents can query the corpus or the live site.1244 PyPI18MIT
- AlicenseNot gradedqualityAmaintenanceA local-first MCP server that indexes Markdown/Obsidian folders and lets AI clients retrieve cited local context through MCP.MIT
- FlicenseBqualityDmaintenanceRead-only MCP server for local Warp/Oz agent data, exposing memories, notebooks, conversations, queries, project rules, and cross-agent handoffs.13-
- AlicenseAqualityDmaintenanceA local-first MCP server that exposes personal notes and files as unified semantic context for AI agents via vector search and file monitoring.6MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.