Skip to content

CHANGELOG.md

Latest commit

 

History

History
178 lines (142 loc) · 8.45 KB

CHANGELOG.md

File metadata and controls

178 lines (142 loc) · 8.45 KB
Raw

Changelog

All notable changes to AI Context Kit are documented here. This changelog helps you decide what to pull into your own instance when the template is updated.

How to use this changelog

When a new version is released:

  1. Check what changed in the version entry below
  2. Copy the files listed under Safe to update from the template into your instance
  3. Never overwrite files listed under Protect — those are your personal artifacts

[Unreleased]

[1.4.3] - 2026-07-07

Fixed

  • Replaced all surviving deterministic phrasing (spec §4.1 footnote, §10.8 workflow table; README session state section) with consistent and predictable — aligns with the reduced-variance language established in v1.4.2
  • Removed stale agents/openai.yaml sidecar claim from README Codex section (sidecars were deleted in v1.4.2 but the description was not updated)
  • Corrected Copilot platform table row: Copilot Code Review does support AGENTS.md; removed incorrect caveat and expanded label to reflect full Copilot coverage
  • Replaced "Works seamlessly" with honest cross-platform framing that matches the platform-variance caveat already present in the same section
  • Added provenance note to agentskills.io link — clarifies it is a third-party community site, not a project-maintained resource
  • Corrected CLAUDE.md / .cursorrules comparison: "No layering" was factually wrong (both support nested files); reframed around the actual gap — no personal/project separation that travels across tools
  • Added deprecation callout to template repository section directing new users to the plugin instead

Safe to update from template

  • specs/context_aware_ai_session_spec.md
  • README.md
  • templates/usercontext_template.instructions.md
  • templates/AGENTS_template.md
  • skills/create-agents-md/SKILL.md
  • skills/create-project-instructions/SKILL.md
  • skills/create-usercontext-instructions/SKILL.md
  • skills/create-usercontext-instructions/references/quality-checklist.md
  • skills/validate-agents-md/SKILL.md
  • skills/validate-usercontext-instructions/SKILL.md
  • skills/validate-project-instructions/SKILL.md

Protect (never overwrite)

  • Your personal *_usercontext.instructions.md
  • Your project AGENTS.md
  • Any custom skills you have created

[1.4.2] - 2026-05-12

Added

  • create-checkpoint skill — guided workflow to capture session state as a checkpoint artifact per spec section 4.4, with schema reference and agents/openai.yaml sidecar
  • restore-checkpoint skill — guided workflow to load, validate, and restore a checkpoint artifact at session start per spec section 4.4 restore rules, with conflict resolution reference and agents/openai.yaml sidecar
  • checkpoint: true required field — added to spec section 4.4.2 schema so checkpoint files are self-identifying; restore validation is now fully spec-backed
  • AGENTS.md now references sections 4.4 (Cross-Session Persistence) and 4.5 (Context Compression) with explicit behavioral instructions for both
  • Prompts and Codex symlinks added for both new skills
  • SKILL.validation.md added to create-checkpoint/ and restore-checkpoint/ skill folders

Safe to update from template

  • skills/create-checkpoint/ (new)
  • skills/restore-checkpoint/ (new)
  • .agents/skills/ (two new symlinks)
  • prompts/ (two new prompt files)
  • specs/context_aware_ai_session_spec.md
  • AGENTS.md
  • README.md
  • CHANGELOG.md

Protect (never overwrite)

  • Your personal *_usercontext.instructions.md
  • Your project AGENTS.md
  • Any custom skills or checkpoint files you have created

[1.4.1] - 2026-05-11

  • Claude Code plugin — .claude-plugin/plugin.json manifest makes the repo installable as a native Claude Code plugin; skills are auto-discovered from skills/
  • GitHub Copilot CLI plugin — .claude-plugin/marketplace.json turns the repo into a self-hosted marketplace, enabling copilot plugin marketplace add MSiccDev/ai-context-kit install flow (Claude Code and Copilot CLI share the same plugin spec)
  • OpenAI Codex support — agents/openai.yaml sidecar added to all 9 skill folders with UI display names, short descriptions, default prompts, and invocation policies for the Codex skill picker
  • .agents/skills/ discovery directory — adds a symlink for each skill folder into .agents/skills/ so Codex auto-discovers skills from its standard path without duplicating content
  • version and allowed-tools fields added to all 9 skills/*/SKILL.md frontmatter
  • README install sections for Claude Code, GitHub Copilot CLI, and OpenAI Codex

Safe to update from template

  • .claude-plugin/ (new directory)
  • .agents/ (new directory)
  • skills/ (all skill folders — updated frontmatter + new agents/openai.yaml in each)
  • templates/skill_template/SKILL.md
  • README.md
  • CHANGELOG.md

Protect (never overwrite)

  • Your personal *_usercontext.instructions.md
  • Your project AGENTS.md
  • Any custom skills you have created

[1.4.0] - 2026-05-08

Added

  • Spec section 4.4 Cross-Session Persistence — normative rules for checkpoint artifacts: proposal rules, required schema (project, role, phase, output_style, tone, interaction_mode, open_tasks, key_decisions, active_files, last_updated), and restore/conflict rules
  • Spec section 4.5 Context Compression — normative rules for managing context window pressure: proposal rules, required compression checkpoint contents, and reversibility requirements
  • docs/spec-rationale.md: new "Cross-Session Persistence and Context Compression" section with example checkpoint file, conflict resolution dialog, and annotated compression proposal dialog

Changed

  • Spec version bumped from 1.3.1 → 1.4.0 across spec, templates, skills, and all stamped artifacts
  • Session Persistence and Context Compression promoted from "Future Enhancements" (rationale doc) to normative spec sections

Safe to update from template

  • specs/context_aware_ai_session_spec.md
  • docs/spec-rationale.md
  • templates/usercontext_template.instructions.md
  • templates/AGENTS_template.md
  • templates/skill_template/SKILL.md
  • skills/ (all skill folders)
  • prompts/ (all prompt files)
  • README.md
  • CHANGELOG.md

Protect (never overwrite)

  • Your personal *_usercontext.instructions.md
  • Your project AGENTS.md
  • Any custom skills you have created under skills/

[1.3.1] - 2026-05-07

Added

  • CHANGELOG.md to track changes and simplify instance update workflow
  • "See It In Action" before/after example section in README
  • "Why This Over Alternatives?" comparison section in README
  • Platform limitations column in the context loading table
  • ## How to Invoke section in all skill SKILL.md files with per-platform guidance
  • Validation scope disclaimer in AGENTS.validation.md, usercontexts/sample_usercontext.validation.md, and README
  • spec_version stamping in templates and enforced in create/validate skill workflows
  • docs/spec-rationale.md companion document (rationale, examples, scenarios)
  • Realistic illustrative sample user context (Jordan Kim persona)

Changed

  • "Determinism" principle renamed to "Reduced variance" throughout spec and README
  • Validate skill Purpose/Constraints clarified to "structured checks based on a deterministic scoring rubric"
  • Platform table: Copilot row updated to distinguish coding agent (AGENTS.md ✅) from Code Review (AGENTS.md ❌)
  • Spec trimmed to normative rules only (426 lines); rationale and examples moved to docs/spec-rationale.md
  • References auto-loading claim scoped to repo-aware environments only

Safe to update from template

  • specs/context_aware_ai_session_spec.md
  • templates/usercontext_template.instructions.md
  • templates/AGENTS_template.md
  • templates/skill_template/SKILL.md
  • skills/ (all skill folders)
  • prompts/ (all prompt files)
  • README.md
  • CHANGELOG.md

Protect (never overwrite)

  • Your personal *_usercontext.instructions.md
  • Your project AGENTS.md
  • Any custom skills you have created under skills/

[1.3.0] - 2025-12-23

Added

  • Initial skill ecosystem: create and validate workflows for user context, project instructions, AGENTS.md, and skills
  • Repository drift-control skill
  • Canonical templates for user context, AGENTS.md, and skill artifacts
  • Session specification v1.3.0

Safe to update from template

  • All files except personal artifacts listed below

Protect (never overwrite)

  • Your personal *_usercontext.instructions.md
  • Your project AGENTS.md