list_consolesmrfentmen/vg-resource-mcp — All available platforms.
download_assetmrfentmen/vg-resource-mcp — Download a single file to disk.
batch_downloadmrfentmen/vg-resource-mcp — Download multiple files to a directory.
check_deprecationtsvillain/deprecation-mcp — Checks whether a vendor API or SDK is active, deprecated, or sunset, returning status and migration details.
generate_image_toolvanshgurawalia/Imaginate — Generates an image from a rough idea by expanding it into a detailed prompt and calling a text-to-image model.
audit_ablationSreeDharshan-GJ/experiment-audit-mcp — Verdict (clean/confounded/uncertain) on whether an ablation pair actually isolates claimed_variable.
audit_sweepSreeDharshan-GJ/experiment-audit-mcp — Hyperparameter importance ranking with a hard 10-run floor and co-varying-parameter warnings.
test_connectionSreeDharshan-GJ/experiment-audit-mcp — Validates W&B credentials; runs automatically on server start, also callable mid-session.
list_runsSreeDharshan-GJ/experiment-audit-mcp — Cheap, paginated run listing (id/name/tags/status only — no config or metrics).
get_run_summarySreeDharshan-GJ/experiment-audit-mcp — Full config + summary metrics + data_completeness for one run.
audit_claimsSreeDharshan-GJ/experiment-audit-mcp — Run the full scientific reasoning pipeline on claims and evidence, returning a structured report.
check_missing_evidenceSreeDharshan-GJ/experiment-audit-mcp — Check which claims lack supporting evidence.
check_scopeSreeDharshan-GJ/experiment-audit-mcp — Check if evidence scope matches claim scope.
check_contradictionsSreeDharshan-GJ/experiment-audit-mcp — Detect contradictions among claims and evidence.
get_metric_historySreeDharshan-GJ/experiment-audit-mcp — Full point-by-point history for one metric on one run.
compare_runsSreeDharshan-GJ/experiment-audit-mcp — Config + metric diff across N runs (not just pairwise). No verdict — just the facts.
audit_training_curveSreeDharshan-GJ/experiment-audit-mcp — Scored signals over a metric history: null_values, sudden_jump, low_variance_plateau, high_frequency_oscillation.
score_confidenceSreeDharshan-GJ/experiment-audit-mcp — Compute confidence scores for claims based on evidence quality and contradictions.
judge_claimsSreeDharshan-GJ/experiment-audit-mcp — Return verdicts (supported/partially supported/unsupported) for each claim.
recommend_actionsSreeDharshan-GJ/experiment-audit-mcp — Generate recommendations for each claim (gather evidence, narrow scope, retract).
generate_reportSreeDharshan-GJ/experiment-audit-mcp — Produce a full scientific report in markdown format from claims and evidence.
calculate_concentrationQINGMINGMIKU/protein-lab — Calculate protein concentration using Beer-Lambert law
plan_dilutionQINGMINGMIKU/protein-lab — Plan a dilution series for BLI or other assays
save_experimentQINGMINGMIKU/protein-lab — Save an experiment record with results and optional raw data
add_proteinQINGMINGMIKU/protein-lab — Add a protein to the local protein library
import_fastaQINGMINGMIKU/protein-lab — Import proteins from a FASTA string
search_proteinsQINGMINGMIKU/protein-lab — Search proteins by name or sequence
export_experimentsQINGMINGMIKU/protein-lab — Export experiments to Excel
read_protein_libraryQINGMINGMIKU/protein-lab — Read the protein library with optional search, tag filter, and sorting
convert_unitsQINGMINGMIKU/protein-lab — Convert concentration between different units
read_research_contextQINGMINGMIKU/protein-lab — Read the research context (goals, subgoals, experiments, conclusions)
rlm_execlogunovFGP/better-rlm — Execute Python code inside the sandboxed REPL to explore large files loaded as variables
rlm_statuslogunovFGP/better-rlm — Report the resolved transport, models selected, and current configuration of the RLM server
rlm_sub_query_batchlogunovFGP/better-rlm — Run a batch of sub-queries (Haiku map-reduce) across chunks of a context.
rlm_sub_querylogunovFGP/better-rlm — Run a single sub-query (Haiku) on a specific context or chunk.
rlm_list_contextslogunovFGP/better-rlm — List all contexts currently loaded in the sandbox.
rlm_load_filelogunovFGP/better-rlm — Load a file from the local filesystem into the sandboxed REPL as a variable.
rlm_chunk_contextlogunovFGP/better-rlm — Split a loaded context into smaller chunks for map-reduce processing.
rlm_greplogunovFGP/better-rlm — Search a loaded context for a pattern using grep-like syntax, returning matching lines. Free, no model call.
rlm_read_chunklogunovFGP/better-rlm — Read a specific chunk of a loaded context by index. Free, no model call.
rlm_drop_contextlogunovFGP/better-rlm — Remove a loaded context from the sandbox to free memory.
rlm_load_contextlogunovFGP/better-rlm — Load a large context (text) into the sandboxed REPL as a variable for exploration without putting it in the prompt.
rlm_querylogunovFGP/better-rlm — Submit a recursive sub-query to explore large content without putting it in the prompt
rlm_inspect_contextlogunovFGP/better-rlm — Inspect a loaded context in the sandbox, e.g., show its size, type, or preview.
execute_postsalmanneedsajob/notary-mcp — Execute a previously approved post revision. Requires the exact revision hash and an operation ID for idempotency.
propose_postsalmanneedsajob/notary-mcp — Propose a post for human approval; returns a draft ID and the canonical hash of the proposed content.
pvr_citiesnotprashanth/pvr-inox-mcp — List cities covered by the PVR/INOX chain
pvr_cinemasnotprashanth/pvr-inox-mcp — List cinemas in a city, including cinema_id
pvr_now_showingnotprashanth/pvr-inox-mcp — Get currently showing films at a cinema
generate_bug_reportAdnan-commits/bug_report_tool — Converts an informal issue description into a structured, developer-ready bug report. Classifies input as bug, feature_request, question, or