Context Bridge
Context Bridge provides a single MCP endpoint to fetch development context from Jira, Sentry, and GitHub Actions for AI agents.
Fetch a Jira ticket by issue key, returning id, summary, status, assignee, and description.
List unresolved Sentry issues for a project slug, with an optional result limit.
List recent GitHub Actions workflow runs for an owner/repo, with an optional result limit.
Requires configured Jira, Sentry, and GitHub credentials via environment variables.
Provides tools for checking GitHub Actions CI/CD run status for a repository.
Provides tools for retrieving Jira ticket data by ID, enabling AI agents to access ticket summaries and statuses.
Provides tools for retrieving Sentry alerts and issues, allowing AI agents to inspect error monitoring context.
Provides tools for sending status updates and notifications to Slack.
Context Bridge
Context Bridge - one MCP endpoint for all your dev tools
Problem:
Tool sprawl across Jira, Sentry, GitHub, and CI systems.
Context fragmentation makes it hard for AI agents to operate across tools.
AI agents are currently blind to critical development context (Jira tickets, Sentry alerts, CI status).
Setup:
pip install -r requirements.txt(or necessary dependencies)Set required environment variables (e.g., API keys for Jira/Sentry).
Add the running MCP server endpoint to your AI environment (e.g., Claude Code or Cursor) as a custom MCP server.
Quick Start
To run the server:
python server.pyEnsure required environment variables are set.
Related MCP server: Aurea Orchestrator MCP
Example
A sample response from get_jira_ticket:
{"id": "PROJ-123", "summary": "Fix login bug", "status": "In Progress"}
Tools:
Tool Name | Description | Status |
| Fetches a Jira ticket by issue key via the Jira REST API. | Yes Implemented |
| Lists unresolved Sentry issues by project slug. | Yes Implemented |
| Checks CI/CD status for a repo. | Yes Implemented |
License: MIT
Related projects
ghost-tracer - local-first observability for AI agents
ghost-reason - trace your agent's reasoning
agent-cost-auditor - audit an agent's log for wasted tokens
Agensi storefront - ready-to-use AI skills for Jira, Sentry, and DevOps
Ready-to-use Agent Skills
Want your AI agent to use these tools automatically? Get the Dev Context Connector skill on Agensi. It includes the exact setup instructions and configuration templates for Claude Code, Cursor, and other MCP clients.
Available Tools
3 toolsget_github_actions_runsARead-only
Lists recent GitHub Actions workflow runs for a repository (owner/repo). Returns JSON array with run id, name, status, conclusion, head_branch, and html_url. Optional limit (default 5). Requires GITHUB_PERSONAL_ACCESS_TOKEN with actions:read scope.
Inputs: repo (required), owner (required), limit. Returns: result
get_jira_ticketARead-onlyIdempotent
Retrieves a Jira ticket by its issue key (e.g. 'PROJ-123'). Returns JSON with id, summary, status, assignee, and description fields. Requires JIRA_BASE_URL, JIRA_EMAIL, and JIRA_API_TOKEN environment variables.
Inputs: ticket_id (required). Returns: result
get_sentry_issuesARead-only
Lists recent unresolved issues from a Sentry project by project slug. Returns JSON array with issue id, title, culprit, count, and first_seen. Optional limit (default 5). Requires SENTRY_ORG_SLUG and SENTRY_AUTH_TOKEN environment variables.
Inputs: project_slug (required), limit. Returns: result
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 3 tools
Each tool targets a completely distinct external service (Sentry, Jira, GitHub Actions), so there is no plausible overlap or misselection risk.
All tools follow a clear get_<service>_<resource> pattern, making intent predictable. Minor inconsistency: get_sentry_issues and get_github_actions_runs are plural while get_jira_ticket is singular, but this reflects the underlying resource semantics.
Three tools is thin but defensible for a read-only context aggregation server where each tool maps to one integration. It is slightly under-scoped rather than bloated.
The surface only supports listing/fetching at one level: no Sentry issue detail, no GitHub run or PR detail, and no Jira search or comment retrieval. Core 'pull context' use cases work, but agents hitting detail or filtering needs will hit dead ends.
Maintenance
Related MCP Connectors
One MCP server for all your work accounts: Slack, Teams, Google Chat, Jira, ClickUp, digest, tasks.
One PAT, any MCP agent: Vercel, GitHub, Cloudflare, Supabase, GCP — unified dev infra gateway.
One MCP endpoint for your company's tools, with per-person sign-in, policy and audit.
MCP server for Sentry - error monitoring, issue tracking, and debugging for AI assistants
Related MCP Servers
- AlicenseAqualityDmaintenanceMCP server for developer workflow automation — smart commits, secret scanning, PR descriptions, and more.5MIT
- AlicenseNot gradedqualityDmaintenanceProduction-ready MCP server for enforcing git/PR/Jira workflows with atomic, composable tools. Enables automated branch creation, conventional commits, PR generation, and Jira ticket operations with safety guardrails.MIT
- AlicenseNot gradedqualityFmaintenanceSelf-hosted MCP gateway that connects productivity tools like Jira, Confluence, and Obsidian to AI assistants via a single endpoint.2MIT
- AlicenseNot gradedqualityFmaintenanceA local, provider-neutral MCP server for repository-scoped issue handling. It provides a guarded interface to Linear, GitHub Issues, GitHub Projects v2, and Jira Cloud, with preview/apply safety and host-local configuration.73 npm1MIT