contexthub-md
Official@contexthub-md
ContextHub skills for mission-driven development: semantic code search, OCXP code graph intelligence, RAG knowledge bases, and transcript-scored mission debriefs.
Agent Skills by contexthub-md
Showing 21 vetted skills indexed across 1 GitHub repositories.
contexthub-mdresearch
Researches topics across project knowledge base, library documentation, and web search in parallel.
contexthub-mdmission-feedback
Debriefs completed missions by analyzing Claude Code session transcripts into scored findings.
contexthub-mddiscover
Searches a project's knowledge base using semantic search and RAG over indexed repos and docs.
contexthub-mdcontext
Read, write, list, and delete ContextHub documents via ch_*_context MCP tools.
contexthub-mdmission-dev
Sets up mission development branches, workspaces, and branch indexing in ContextHub.
contexthub-mdkb
Query project reference documentation repos using semantic search and RAG answers.
contexthub-mdplan
Generates implementation plans for features and refactors using OCXP code graph intelligence.
contexthub-mdsop
Create, list, and read Standard Operating Procedures stored as markdown in ContextHub projects.
contexthub-mdresume
Loads previous ContextHub session context and proposes next steps to continue work.
contexthub-mdreflect
Captures post-run product feedback and process lessons into dated markdown reflection files.
contexthub-mdstart
Maps user intents to the correct ContextHub ch_* tool or skill.
contexthub-mdstatus
Retrieves project health briefings covering active missions, open memos, and code index status.
contexthub-mdsop-writer
Creates validated standard operating procedures grounded in codebase research and OCXP code intelligence.
contexthub-mdmission
Manages development missions in ContextHub by listing, creating, updating, and tracking tasks.
contexthub-mdexplore
Navigate and understand codebases using OCXP code graph intelligence and semantic search.
contexthub-mdanalyze
Audit code quality, dead code, and architecture using OCXP graph intelligence.
contexthub-mdbranch-search
Indexes and searches Git branches using ContextHub OCXP code intelligence.
contexthub-mdmission-forge
Converts feature ideas into executable ContextHub missions with validated specs, phases, and agent assignments.
contexthub-mdconnect
Configures workspace, project, and mission context for ContextHub MCP tools.
contexthub-mdimport
Imports a GitHub repository into ContextHub and triggers code intelligence indexing.
contexthub-mdweaver-plan
Author goal-focused Weaver Hub Manager-mode mission plans and land them via Brain or local YAML.
Frequently Asked Questions About contexthub-md
FAQPage SchemaWhat tasks can I accomplish with contexthub-md skills?▼
You can explore and audit codebases with OCXP code graph intelligence, run semantic RAG search over project knowledge bases, create and track development missions, author validated SOPs and runbooks, index and grep Git branches, and debrief completed missions from their actual session transcripts with scored findings.
Who are the contexthub-md skills designed for?▼
They target software engineers and technical leads working in ContextHub projects who need code navigation, blast-radius impact analysis before refactors, mission planning and tracking, and evidence-based post-run reviews. Teams dogfooding work through Weaver Hub over MQTT also get dedicated Manager-mode mission plan authoring.
How do I get started with a ContextHub project using these skills?▼
Run the connect skill first for guided workspace, project, and mission setup, then use import to add a GitHub repository, which downloads the code and triggers code intelligence indexing. The start skill provides an orientation map of which ch_* tool or /contexthub skill fits each intent.
What external services and dependencies do these skills rely on?▼
Skills depend on ContextHub with OCXP code graph intelligence for search, planning, and analysis. The research skill additionally uses Context7 for external library documentation and Tavily for web search. Weaver Hub mission plans are delivered over MQTT, and mission-feedback reads Claude Code session transcripts.
How does mission-feedback differ from the reflect skill?▼
Mission-feedback is transcript-driven and scored: it resolves a mission ID to its actual session transcript, reads prompts, tool calls, errors, and token usage, then produces ranked findings classified as bugfix, prompt-change, or skill-creation. Reflect is a subjective, manual post-run reflection saved as a dated markdown feedback file.