Skip to content

Claude/analyze repo improvements lnt sd - #4

Closed
hyperpolymath wants to merge 4 commits into
mainfrom
claude/analyze-repo-improvements-LntSD
Closed

Claude/analyze repo improvements lnt sd#4
hyperpolymath wants to merge 4 commits into
mainfrom
claude/analyze-repo-improvements-LntSD

Conversation

@hyperpolymath

@hyperpolymath hyperpolymath commented Apr 9, 2026

Copy link
Copy Markdown
Owner

Closed — all changes already merged to main

This PR's changes were merged to main via:

The branch diverged from main after PR #2 made independent enhancements to llm_client.jl (structured error handling, provider validation), creating irreconcilable merge conflicts in the branch history. Since all substantive work is on main, closing this PR rather than force-resolving stale conflicts.

claude added 2 commits April 9, 2026 10:06
- Create engine/src/metrics.jl: implements compute_metrics, MetricsSnapshot,
  metrics_summary, and supporting functions (vocabulary diversity, reference
  rates, coherence, convergence, novel n-grams). This was a critical missing
  file that prevented the JeffEngine module from loading.
- Rewrite thejeffparadox_README.md: replace incorrect npm/pip placeholder
  with actual Julia/Hugo/Ada instructions matching the real project.
- Fix rate limiting in llm_client.jl: implement token-based rate limiting
  that was previously stubbed out, and properly synchronize cleanup of
  request_times and token_counts vectors.
- Improve Ada YAML parsing in game_state.adb: add line trimming, comment
  skipping, key-at-start-of-line matching, and quoted string handling for
  current_node values.

https://claude.ai/code/session_01FdPFGKmDfdZYBgi2jGGoAv
- Split llm_client.jl into three focused modules:
  - llm_client.jl: HTTP transport, context building, generation
  - embedding_client.jl: embedding generation, cosine similarity
  - rate_limiter.jl: per-provider rate/token limiting
- Wire up compute_aperture() in generate_response() via keyword arg
  `game`, so temperature is dynamically adjusted based on conversation
  metrics (vocabulary diversity, coherence). Previously this was a TODO.
- Extract hardcoded model names into DEFAULT_MODELS config dict,
  overridable via GM_MODEL env var.
- Add comprehensive tests: state persistence round-trip, rate limiter
  token enforcement, cosine similarity, coherence/convergence metrics,
  LLM client header/body/response parsing, and metrics summary.
- Fix CONTRIBUTING.md: replace stale nix develop/flake.nix/just check
  references with actual Julia/Hugo setup instructions and correct
  repository structure.
- Fix Containerfile: replace Ada TUI placeholder comment with actual
  GNAT build attempt that gracefully skips when toolchain unavailable.

https://claude.ai/code/session_01FdPFGKmDfdZYBgi2jGGoAv
@github-actions github-actions Bot added documentation Docs, prose, diagrams, READMEs, ADRs engine labels Apr 9, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

documentation Docs, prose, diagrams, READMEs, ADRs engine

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants