feat: add soldexter ECC bundle

[ecc-tools]

Summary

Auto-generated ECC bundle from repository analysis.

What This Does

Merging this PR adds repo-local ECC artifacts for both Claude Code and Codex. The generated bundle captures repository patterns, Codex baseline config, and reusable workflow scaffolds derived from git history analysis.

Analysis Scope

• Commit history patterns and conventions
• Code architecture and structure
• Testing patterns and coverage
• Recurring workflows

Files

Path	Description
.claude/ecc-tools.json	ECC install manifest used for upgrades, repair, and uninstall.
.claude/skills/soldexter/SKILL.md	Repository-specific Claude Code skill generated from git history.
.agents/skills/soldexter/SKILL.md	Codex-facing copy of the generated repository skill.
.agents/skills/soldexter/agents/openai.yaml	Codex skill metadata so the repo skill appears cleanly in the skill interface.
.claude/identity.json	Suggested identity.json baseline derived from repository conventions.
.codex/config.toml	Repo-local Codex MCP and multi-agent baseline aligned with ECC defaults.
.codex/AGENTS.md	Codex usage guide that points at the generated repo skill and workflow bundle.
.codex/agents/explorer.toml	Read-only explorer role config for Codex multi-agent work.
.codex/agents/reviewer.toml	Read-only reviewer role config focused on correctness and security.
.codex/agents/docs-researcher.toml	Read-only docs researcher role config for API verification.
.claude/homunculus/instincts/inherited/soldexter-instincts.yaml	Continuous-learning instincts derived from repository patterns.

Optional: Continuous Learning (15 instincts)

This PR also includes instincts for the continuous-learning-v2 skill. These are optional and only useful if you use that skill.

Import after merging:

/instinct-import .claude/homunculus/instincts/inherited/soldexter-instincts.yaml

Review Checklist

• Verify detected patterns are accurate
• Confirm generated config, commands, and skill metadata match the repo’s real workflow
• Check best practices align with team standards before merging

---

_{ECC Tools | Everything Claude Code}

Summary by Sourcery

Add repository-local ECC configuration, skills, and continuous-learning instincts for the soldexter project to support Claude Code and Codex workflows.

New Features:

• Introduce soldexter-specific development skill guides for both Claude and Codex, documenting naming conventions, commit patterns, and contribution workflows.
• Add Codex configuration and multi-agent role definitions (explorer, reviewer, docs researcher) tailored to this repository.
• Provide Codex/OpenAI skill metadata so the soldexter repo skill appears in the Codex skill interface.
• Include continuous-learning instincts derived from soldexter repository patterns for use with the Claude homunculus system.

Documentation:

• Document soldexter development patterns, testing practices, and contributor workflows in repo-local SKILL and AGENTS guides.

Chores:

• Add ECC manifest and identity baseline files to integrate this repository with ECC tools and conventions.

[sourcery-ai]

Reviewer's Guide

Adds an auto-generated ECC bundle for the soldexter repo, including Claude and Codex skills describing repo conventions, Codex multi-agent and MCP config, and optional continuous-learning instincts that encode detected commit, naming, import/export, and testing patterns.

Sequence diagram for importing soldexter continuous-learning instincts

sequenceDiagram
  actor Developer
  participant ClaudeCode
  participant Homunculus

  Developer->>ClaudeCode: /instinct-import .claude/homunculus/instincts/inherited/soldexter-instincts.yaml
  ClaudeCode->>Homunculus: /instinct-import .claude/homunculus/instincts/inherited/soldexter-instincts.yaml

Loading

File-Level Changes

Change	Details	Files
Introduce continuous-learning instincts capturing soldexter git, naming, and testing conventions.	Add multiple instincts with ids for commit format/length, file/function/class/constant naming, import/export style, and test placement. Document each instinct with trigger phrases, confidence, domain, and evidence based on repository analysis. Mark the instincts as supplemental to existing curated instincts and provide import guidance in the PR description.	.claude/homunculus/instincts/inherited/soldexter-instincts.yaml
Add mirrored soldexter development-pattern skills for Claude Code and Codex.	Create a repo-specific skill document describing file naming, import/export styles, commit conventions, workflows, and testing patterns. Define two main workflows (/update-docs and /contribute-code) with step-by-step guidance. Mirror the same content for Claude (.claude/skills/...) and Codex (.agents/skills/...) so both systems share a consistent skill view.	.claude/skills/soldexter/SKILL.md .agents/skills/soldexter/SKILL.md
Configure Codex MCP servers and multi-agent baseline tailored to ECC defaults.	Add a Codex config enabling GitHub, Context7, Exa, Memory, Playwright, and Sequential Thinking MCP servers with appropriate commands/URLs. Enable multi_agent support with thread and depth limits and register explorer, reviewer, and docs_researcher agents pointing to per-agent config files. Set top-level policies such as approval_policy, sandbox_mode, and web_search behavior for this repo.	.codex/config.toml
Define Codex multi-agent role configs and documentation for repo-local ECC usage.	Add explorer, reviewer, and docs-researcher agent configs that specify model, sandbox_mode, and tightly scoped developer instructions for each role. Document how the repo-local ECC setup integrates with Codex CLI, including where the repo skill lives and what MCPs/agents are available. Clarify that no workflow command files were generated and these agents/skills act as reusable scaffolds aligned with detected workflows.	.codex/agents/explorer.toml .codex/agents/reviewer.toml .codex/agents/docs-researcher.toml .codex/AGENTS.md
Expose the soldexter repo skill in the Codex skill interface with safe defaults.	Add OpenAI agent metadata pointing to the soldexter skill with display name, short description, and a default prompt that nudges use of repo conventions. Allow implicit invocation so the skill can be auto-suggested in relevant contexts.	.agents/skills/soldexter/agents/openai.yaml
Add ECC manifest and Claude identity baseline files for this repo.	Introduce an ECC tools manifest used for installation, upgrades, and repair of the Claude-side ECC bundle. Add a suggested Claude identity.json baseline consistent with conventions inferred from the repository.	.claude/ecc-tools.json .claude/identity.json

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[sourcery-ai]

Hey - I've left some high level feedback:

• The SKILL.md files under .agents/skills/soldexter/ and .claude/skills/soldexter/ are entirely wrapped in a ```markdown code fence, which will render the whole document as a code block rather than normal markdown; consider removing the outer fence so the content formats correctly.
• Several instincts in .claude/homunculus/instincts/inherited/soldexter-instincts.yaml appear to be duplicated or highly overlapping (e.g., separate entries for file naming, import/export style, and commit length/format), so it may be worth consolidating or de-duplicating these to avoid redundant guidance.
• Some of the commit- and naming-related instincts explicitly note they are based on a single commit or limited evidence but still use relatively high confidence values; you might want to lower those confidences or soften the actions to reflect that the patterns are not yet well established.

Prompt for AI Agents

Please address the comments from this code review:

## Overall Comments
- The SKILL.md files under `.agents/skills/soldexter/` and `.claude/skills/soldexter/` are entirely wrapped in a ```markdown code fence, which will render the whole document as a code block rather than normal markdown; consider removing the outer fence so the content formats correctly.
- Several instincts in `.claude/homunculus/instincts/inherited/soldexter-instincts.yaml` appear to be duplicated or highly overlapping (e.g., separate entries for file naming, import/export style, and commit length/format), so it may be worth consolidating or de-duplicating these to avoid redundant guidance.
- Some of the commit- and naming-related instincts explicitly note they are based on a single commit or limited evidence but still use relatively high confidence values; you might want to lower those confidences or soften the actions to reflect that the patterns are not yet well established.

---

Sourcery is free for open source - if you like our reviews please consider sharing them ✨

• X
• Mastodon
• LinkedIn
• Facebook

_{Help me be more useful! Please click 👍 or 👎 on each comment and I'll use the feedback to improve your reviews.}