claudecodex

project-mining

Mine project histories for behavioral evidence. Search chat logs, git history, docs, and code for concrete examples of specific patterns, values, or skills.

1 on GitHubby Cory Kingv3.0.0MITupdated 20 days ago
/

Plugin project-mining from marketplace coryking-plugins in coryking/coryking-plugins. Commands read from the repo's manifest — 20 days ago.

This plugin runs code on your machine — it starts local MCP servers. The exact commands are listed below; read them before installing.

What it installs

Skills (3)

  • activity-reflection

    Analyse how human attention was spent across agent sessions over a window, combining activity timeline proxies with original conversation evidence. Use for attention, delegation, intervention and recurring activity reports.

  • cc-explorer

    Search and inspect Claude Code and Codex history for decisions, original wording, agent prompts, tool actions, and usage. Use for questions about past conversations or agent runs.

  • project-mining

    Build an evidence-backed account of a project's work against a question, rubric or lens, combining history, source and actual outputs. Use for portfolio evidence, patterns, skills and cross-source investigations; quick history lookups use cc-explorer.

Subagents (2)

  • activity-analyst

    Interpret human attention across agent sessions using timeline measurements and original conversation evidence, with calibrated sampling and explicit limits.

  • evidence-worker

    Find relevant original evidence across conversation history, repository artifacts and preserved outputs; return excerpts and locators for the caller to judge.

MCP servers (1)

  • cc-explorer

    runs: /bin/sh -c exec env UV_CACHE_DIR="${TMPDIR:-/tmp}/cc-explorer/uv-cache" UV_PROJECT_ENVIRONMENT="${TMPDIR:-/tmp}/cc-explorer/.venv" uv run --project . cc-explorer

View source on GitHub

More from coryking/coryking-plugins

All 4 plugins →

engineering-loop

claudecodex

Slim parallel-review and curated research agents for solo development. The parts of compound-engineering's workflow that actually compound for a single operator, without the team-scale scaffolding.

3 skills · 20 agents

1

squabble

claude

A panel of expert reviewers that argues over the design of a tool, model, spec, plan, or decision while it is still cheap to change — and hands back one record of what the panel agreed on, disputed, and withdrew.

1 skill · 9 agents

1

Reference guidance for writing MCP tool descriptions, server instructions, parameter schemas, output schemas, and annotations that LLMs actually use well.

1 skill

1

Similar plugins

让 Claude 在本地读懂你的微信:查询、导出微信 4.1+ 聊天记录为 AI 可读文本。

2 skills

chengmarc/wechat-to-ai 41

Exports current session segment (since last compaction) with system-reminder stripping -- main conversation, subagent transcripts, sidechains, and debug logs in structured markdown

hooks

Emasoft/emasoft-chat-history 5

Maintain project-mining?

This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.

claude & codex plugins4 plugins