claude

code-archaeology-analyst

Autonomous codebase archaeology agent. Triggered by a GitHub Issue or Azure DevOps Work Item tagged 'code-archaeology'. Segments the codebase, maps architecture in business language, extracts coding conventions, runs a due diligence audit for defects and structural problems, classifies all work into typed and prioritised items, and delivers a single structured archaeology report as comments on the originating issue or work item.

1 on GitHubby Xianixv2.0.0MITupdated 10 days ago

Not in a marketplace yet, so it loads for one session with --plugin-dir.

Plugin code-archaeology-analyst in xianix-team/plugins-official. Commands read from the repo's manifest — 10 days ago.

This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.

What it installs

Skills (7)

  • audit-codebase

    Run only Phase 4 — due diligence audit for a single segment. Actively searches for logic defects, design violations, security gaps, fragile patterns, test blind spots, and consistency breaks. Assigns Critical / High / Medium / Low severity with Fix-in-place / Quarantine / Encode-as-prohibition recom

  • classify-debt

    Run only Phase 5 — debt and gap classification. Takes architecture maps, pattern results, and due diligence findings and classifies all work into Enhancement / Remediation / Migration at P1–P5 priority. Produces the prioritized work backlog and summary. Usage: /classify-debt

  • extract-patterns

    Run only Phase 3 — pattern extraction for a single segment. Reads 10–20 representative files and extracts eight pattern types (naming, error handling, ORM/DB, API shape, auth, testing, dependency wiring, configuration) with Consistent / Inconsistent / Split verdicts. Usage: /extract-patterns <segmen

  • map-architecture

    Run only Phase 2 — architecture mapping for a single segment. Reads every source file in the segment and produces module descriptions (Module/Purpose/Owns/Calls/Exposes), service boundary rules, integration points, and data flow traces. Usage: /map-architecture <segment-path>

  • post-report

    Post a completed archaeology report as ordered comments on a GitHub Issue or Azure DevOps Work Item and apply the 'archaeology-complete' label/tag. Usage: /post-report [issue <n> | wi <id>]

  • run-archaeology

    Run the full autonomous codebase archaeology analysis on a GitHub Issue or Azure DevOps Work Item tagged 'code-archaeology'. Segments the codebase, maps architecture, extracts conventions, audits for defects, classifies all work, and posts the complete report as ordered comments on the originating i

  • segment-codebase

    Run only Phase 1 — autonomous segmentation. Reads the codebase structure and divides it into analysis-ready segments without asking. Applies the 15–25 file size limit, cohesion grouping, and confidence self-assessment. Usage: /segment-codebase [path]

Slash commands (1)

  • /code-archaeology-analyst:code-archaeology

    Autonomous codebase archaeology analysis triggered by a GitHub Issue or Azure DevOps Work Item tagged 'code-archaeology'. Fetches the item, runs five autonomous phases across all segments, and delivers the complete report as ordered comments on the originating issue — plus writing the full report to

Subagents (7)

  • architecture-mapper

    Per-segment architecture mapper (Phase 2). Reads every source file within a segment boundary and produces business-language module descriptions, service boundary rules, integration points, and data flow traces. Uses the Module/Purpose/Owns/Calls/Exposes format defined in the archaeology protocol.

  • debt-classifier

    Cross-segment debt and gap classifier (Phase 5). Takes all Phase 2–4 outputs across every segment and classifies every identified work item into Enhancement, Remediation, or Migration — with P1–P5 priority. Applies disambiguation rules autonomously. Produces the full prioritized work backlog and the

  • due-diligence-auditor

    Per-segment due diligence auditor (Phase 4). Actively searches for defects and structural problems that new code could inherit — logic defects, design violations, security gaps, fragile patterns, test blind spots, and consistency breaks. Assigns Critical / High / Medium / Low severity and a Fix-in-p

  • orchestrator

    Code archaeology orchestrator. Detects the hosting platform, fetches the triggering issue or work item, posts an "Analysis in Progress" comment, runs five autonomous phases (Segmentation → Architecture Mapping → Pattern Extraction → Due Diligence Audit → Debt Classification), compiles the full archa

  • pattern-extractor

    Per-segment pattern extractor (Phase 3). Reads 10–20 representative files across a segment and extracts eight pattern types — naming, error handling, ORM/DB access, API response shape, authentication, testing, dependency wiring, and configuration. Assigns Consistent / Inconsistent / Split verdicts.

  • report-writer

    Archaeology report writer. Compiles all phase outputs into the single structured Codebase Archaeology Report defined in the protocol. Writes to ai-dlc/reports/code-archaeology-analysis.md. Includes the Autonomous Decision Log and confidence assessment. Does not stream findings progressively — delive

  • segmentation-analyst

    Autonomous codebase segmentation analyst. Reads the top-level directory structure and decides how to divide the codebase into analysable segments without asking. Applies size limits, cohesion grouping, and confidence self-assessment. Produces the segment plan that all subsequent per-segment agents w

Hooks (2)

  • PreToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/validate-prerequisites.sh

  • PostToolUse

    bash ${CLAUDE_PLUGIN_ROOT}/hooks/notify-archaeology.sh

View source on GitHub xianix.com

More from xianix-team/plugins-official

All 20 plugins →

AI-native UX mob elaboration workflow for Greenfield and Brownfield projects. Runs structured, human-gated phase processes with artifact guardrails, approval gates, context isolation, and AI readiness checks before any brownfield work begins.

3 skills · 9 commands · 6 agents · hooks

1

Automated dependency audit and optimization across all major ecosystems (npm, pip, Cargo, Go modules, Maven, .NET, Bundler). Runs four parallel specialized analyses — vulnerability scanning (CVE/CVSS), SemVer version drift, bundle bloat and unused package detection, and license compliance. Automatically creates a pull request with safe fixes applied; flags critical CVEs and major breaking upgrades for manual review. Works with GitHub, Azure DevOps, and any git repository.

6 skills · 1 command · 5 agents · hooks

1

Automated impact analysis and risk-based test strategy generation for bug fixes, PBIs, and feature implementations. Gathers work item requirements, code changes across linked pull requests, comments, child items, and referenced documentation — then posts a business-readable Markdown comment series on the PR / issue / work item discussion that serves as the testing guide for QA engineers doing risk-based testing. No HTML files written to the repo.

6 skills · 1 command · 5 agents · hooks

1

Comprehensive PR review plugin with specialized agents for code quality, security, test coverage, and performance analysis. Works with GitHub, Azure DevOps, Bitbucket, and any git repository.

6 skills · 1 command · 4 agents · hooks

1

Authorized web-application penetration testing with whitebox source-code analysis. Combines OWASP Top 10 dynamic probes (DAST), static code analysis (SAST via semgrep), web-ecosystem dependency CVE scanning, and secrets detection. Source-code recon runs first and drives targeted URL probing — including authenticated routes, real parameter names, and sink locations — instead of blind crawling. Includes a Playwright-driven probe path for SPA targets and DAST+SAST dual-confirmation. Requires explicit --authorized flag.

1 command · 9 agents · hooks

1

Unified impact analysis and risk-based test strategy plugin. Combines blast radius analysis, dependency tracing, feature mapping, and structured test case generation into a single HTML report. Works from a PR, GitHub issue, or Azure DevOps work item.

1 skill · 1 command · 7 agents · hooks

1

Similar plugins

AI-powered codebase understanding — analyze, visualize, and explain any project

9 skills · 10 agents · hooks

Egonex-AI/Understand-Anything 92,410

Repomix MCP server for AI-powered codebase analysis. Pack local/remote repositories and search or read the generated outputs. Foundation plugin that enables all Repomix features in Claude Code.

1 MCP server

yamadashy/repomix 29,255

Assess repository readiness for AI coding agents with an evidence-based scorecard and prioritized fixes.

2 skills

hashgraph-online/awesome-codex-plugins 1,566

Analyze codebases to generate structured technical documentation with source-of-truth citations. Optimized for AWS workloads using CDK, CloudFormation, and Terraform.

1 skill · 2 MCP servers

awslabs/agent-plugins 941

Periodic full codebase review using automatic-code-review rules. Reviews entire codebase, not just modified files.

1 command

NTCoding/claude-skillz 364

tome

claude

Multi-source research: code archaeology, community discourse (HN, Reddit), academic literature (arXiv, Semantic Scholar), and TRIZ cross-domain analysis

10 skills · 4 commands · 6 agents · hooks

athola/claude-night-market 350

Maintain code-archaeology-analyst?

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 plugins20 plugins