v0.3.5

What's fixed

• Fixed GitHub Copilot context cost accounting to include prompt files, nested AGENTS.md files, and all nested .github/instructions/**/*.instructions.md files.
• Added Copilot MCP config discovery for .vscode/mcp.json and .github/mcp.json, including tools allowlists.
• Updated docs for Copilot instructions, prompts, skills, and MCP cost coverage.

Validation

• npm test
• npm run build

v0.3.4 - context cost estimator

v0.3.4 - context cost estimator

skill-doctor v0.3.4 adds a local context-cost check for AI agent instruction layers.

Highlights

• Adds skill-doctor cost and skill-doctor context as aliases for estimating per-turn context token tax.
• Grades the estimate against a configurable token budget.
• Adds --fail-on-budget so teams can use the check in CI before always-on instructions grow too large.
• Adds JSON output for automation and dashboards.

Why it matters

Agent setups now often include Claude skills, Cursor rules, Copilot instructions, Codex AGENTS.md, and other always-on files. Even when those files are safe, they can quietly add context cost to every turn.

The new command gives users a concrete number and a short recommendation for the highest-cost items.

Try it

npx @evilstar2025/skill-doctor cost
npx @evilstar2025/skill-doctor cost --budget-tokens 2000 --fail-on-budget
npx @evilstar2025/skill-doctor context --json

Safe demo:

git clone https://github.com/evilstar2016/skill-doctor.git
cd skill-doctor/examples/conflicted-agent-project
npx @evilstar2025/skill-doctor scan --scope project
npx @evilstar2025/skill-doctor conflicts --scope project
npx @evilstar2025/skill-doctor audit --scope project
npx @evilstar2025/skill-doctor cost --scope project

Feedback wanted

I am looking for real, redacted examples of:

• always-on agent instruction files that feel too expensive for every turn
• false positives or misleading recommendations from skill-doctor cost
• missing agent platform paths or instruction file patterns
• cases where overlapping AI agent rules made a tool behave inconsistently

Actionable reports: #4

Lightweight questions and examples: #6

The project reads local files and reports findings. It does not upload your skill files by default.

Verification

• npm run build passed.
• npm test passed: 39 test files, 329 tests.
• Official npm latest is @evilstar2025/skill-doctor@0.3.4.

v0.3.3 - npm discovery metadata

v0.3.3 - npm discovery metadata

skill-doctor v0.3.3 is a metadata-only patch release focused on npm discovery.

Highlights

• Adds npm package keywords for AI agent, CLI, developer tooling, and security audit discovery.
• Keeps runtime behavior unchanged from v0.3.2.
• Keeps the safe demo, README quickstart path, and contribution/feedback surfaces from v0.3.2.

Try it

npx @evilstar2025/skill-doctor scan

Safe demo:

git clone https://github.com/evilstar2016/skill-doctor.git
cd skill-doctor/examples/conflicted-agent-project
npx @evilstar2025/skill-doctor scan --scope project
npx @evilstar2025/skill-doctor conflicts --scope project
npx @evilstar2025/skill-doctor audit --scope project

Expected demo output: https://github.com/evilstar2016/skill-doctor/blob/main/docs/demo-output.md

Comparison page: https://github.com/evilstar2016/skill-doctor/blob/main/docs/comparisons/manual-agent-config-audit.md

Feedback wanted

I am looking for real, redacted examples of:

• false positives from skill-doctor conflicts or skill-doctor audit
• missing agent platform paths or instruction file patterns
• risky instruction patterns that should be detected
• cases where overlapping AI agent rules made a tool behave inconsistently

Actionable reports: #4

Lightweight questions and examples: #6

The project reads local files and reports findings. It does not upload your skill files by default.

Verification

• npm run build passed.
• npm test passed: 38 test files, 321 tests.
• npm pack --dry-run included README, assets, examples, and dist files.
• Published to npm as @evilstar2025/skill-doctor@0.3.3.

v0.3.2 — Quickstart demo and contributor launch kit

Highlights

• Adds a runnable safe demo project under examples/conflicted-agent-project so new users can see scan, conflict, and audit findings without scanning their own setup first.
• Adds README quickstart demo artwork and clearer 30-second trial path.
• Adds CHANGELOG.md, ROADMAP.md, CONTRIBUTING.md, issue templates, pull request template, and release checklist.
• Updates npm package metadata and packaged files so README assets and examples ship with the package.

Try it

npx @evilstar2025/skill-doctor scan

Safe demo:

git clone https://github.com/evilstar2016/skill-doctor.git
cd skill-doctor/examples/conflicted-agent-project
npx @evilstar2025/skill-doctor scan --scope project
npx @evilstar2025/skill-doctor conflicts --scope project
npx @evilstar2025/skill-doctor audit --scope project

Expected demo output: https://github.com/evilstar2016/skill-doctor/blob/main/docs/demo-output.md

Feedback wanted

I am looking for real, redacted examples of:

• false positives from skill-doctor conflicts or skill-doctor audit
• missing agent platform paths or instruction file patterns
• risky instruction patterns that should be detected
• cases where overlapping AI agent rules made a tool behave inconsistently

Please report feedback here: #4

For lightweight questions and examples before filing an issue, use Discussion #6: #6

The project reads local files and reports findings. It does not upload your skill files by default.

Launch materials

• First-wave distribution queue: https://github.com/evilstar2016/skill-doctor/blob/main/marketing/distribution/2026-06-21-first-wave-queue.md
• Launch kit: https://github.com/evilstar2016/skill-doctor/blob/main/marketing/launch-kit.md

Verification

• npm test passed: 38 test files, 321 tests.
• npm run build passed.
• npm pack --dry-run included README, assets, examples, and dist files.
• Published to npm as @evilstar2025/skill-doctor@0.3.2.

v0.3.0 — Dashboard + v0.2.0 improvements

What's new in v0.3.0

Dashboard command

New skill-doctor dashboard command generates a unified Mission Control–style HTML dashboard combining all diagnostics — scan, conflicts, audit, and cleanup — in a single scrolling page.

skill-doctor dashboard                           # writes skill-doctor-dashboard.html
skill-doctor dashboard --report ./out/dash.html  # custom output path
skill-doctor dashboard --open                    # open in browser after generating
skill-doctor dashboard --scope project           # project skills only

Features:

• Health ring — SVG donut chart showing proportion of clean / conflicting / at-risk / duplicate skills
• Platform distribution — horizontal bar chart per platform
• Skill inventory — full table with status indicators
• Conflicts section — severity distribution bar + conflict pair cards with similarity scores
• Security audit — 4-rule heatmap (shell-exec, destructive, secret-leak, network-call) + finding detail cards with code snippets
• Cleanup section — duplicate skill pairs with keep/remove recommendations

---

What was new in v0.2.0

• diff command — compare two skills side-by-side with optional LLM analysis
• cleanup command — detect and remove duplicate skills
• Redesigned HTML reports — gstack-decoder aesthetic with dark mode, expandable conflict cards
• OpenClaw & Hermes platform support
• paths.extra config — add custom scan directories with glob wildcard support
• Configurable LLM timeout via config.json

v0.1.0 — Multi-platform Skill Scanner

What's in v0.1.0

Multi-platform AI skill scanner and conflict detector for local CLI use.

Features

• Scan skills across 9 platforms: Claude, Cursor, Copilot, Codex, Gemini, Windsurf, Trae, OpenCode, Kiro
• Detect conflicts using Jaccard similarity with high/med/low severity grading
• Three commands: `scan`, `show `, `conflicts`
• Filters: `--scope`, `--kind`, `--limit`, `--fail-on`, `--json`
• 95%+ test coverage on core modules, 34 tests passing
• Cold start ~1.3s

Install

git clone https://github.com/evilstar2016/skill-doctor.git
cd skill-doctor
npm install && npm run build && npm link
skill-doctor scan

Status

Private dogfood release. Not yet published to npm.