Investigate Ai Session
by amplitude96fc7d4c58bbNo license42 starsListed Oct 8, 2026Updated Oct 8, 2026Repository updated today
Deep-dives into specific AI agent sessions or failure patterns to explain why something went wrong. Only use when the user has Amplitude Agent Analytics instrumented in their project. Use when investigating a specific session ID, debugging agent failures, understanding why quality is low, tracing tool errors, or when monitor-ai-quality surfaces an issue that needs root cause analysis.
- 96fc7d4c58bbCurrentcommit 96fc7d4Published Oct 8, 2026
Source and attribution
Source:amplitude/mcp-marketplaceinplugins/amplitude/skills/investigate-ai-sessionat commit96fc7d4
License: No license
Content belongs to its original authors. SourceWeft indexes it from a public repository.
More from amplitude/mcp-marketplace
What Would Lenny Do
amplitude
Answers product strategy, growth, pricing, hiring and leadership questions by searching Lenny Rachitsky's archive and synthesizing a recommendation.
Weekly Brief
amplitude
Produces a weekly Amplitude analytics briefing covering trends, wins, risks, and next-week priorities.
User Cohort Forensics
amplitude
Investigates individual users and user populations in Amplitude, resolving identities, reading profiles and analyzing event timelines.
Use Amp Guides Surveys
amplitude
Lists Amplitude Guides and Surveys or reads one item's full configuration via the use_amp_guides_surveys tool.
Taxonomy
amplitude
Reference guidance for Amplitude event taxonomy: naming standards, data quality auditing, scoring, and safe deprecation.
Scheduled Report Refresh
amplitude
Refreshes recurring reports (daily briefings, WBRs, experiment briefs) from saved Amplitude content — fetch the dashboard, re-pull its charts, fill gaps with ad hoc queries, summarize deltas. Use for "run the daily/weekly report", "refresh this dashboard's numbers", or scheduled custom-agent report runs.
More in AI & Agents
Using Superpowers
obra
Establishes a rule that agents must invoke relevant skills before any response, with per-platform reference files.
Dispatching Parallel Agents
obra
A pattern for splitting independent tasks across parallel subagents with isolated context, then reviewing and integrating their results.
Adobe
adobe
Renders an informational onboarding tour of the Adobe for creativity connector and its available capabilities.
Show Airtable Link
Airtable
Formats a clickable Airtable link after MCP calls that touch Airtable content, using only IDs returned by tools.
Airtable Filters
Airtable
Builds Airtable filter parameters for MCP tools that list or display records, using field-type-aware operators and nested logic.
Agent Activity Log
Airtable
Scaffolds and operates an opt-in Airtable audit log table recording an agent's decisions, blockers and outcomes.