[skills] Fix auto-capture-claude-code parser and add ingest size guard

[eazene]

Contribution Type

• Repo improvement (bug fix to an existing skill)

What does this do?

Fixes two bugs in skills/auto-capture-claude-code/session-end-capture.mjs that made the reference hook capture nothing against real Claude Code sessions:

1. Transcript parser — parseTranscript expected a plaintext Human:/Assistant: format, but Claude Code writes transcripts as JSONL (one type-tagged JSON object per line). The plaintext parser matched zero turns on every real transcript → the hook always logged skipped:too_short and captured nothing. Rewrote it to parse JSONL: extract type:user/assistant turns, exclude tool-result messages and sidechain (subagent) turns.
2. No payload size guard — long sessions exceed the ingest embedding model's ~8191-token limit, so /ingest returned HTTP 500 (and, because the endpoint upserts in parallel with the embedding call, left an embedding-less row). Added a 24k-char cap (OB_CAPTURE_MAX_CHARS) that keeps the session start and end and drops the middle with a marker.

Only session-end-capture.mjs + a version bump in metadata.json. No behavior change to any Edge Function.

Requirements

Node.js 18+, a deployed Open Brain (open-brain-rest /ingest + MCP_ACCESS_KEY). No new dependencies.

Checklist

• I've read CONTRIBUTING.md
• Tested on my own Open Brain instance (real 16-turn transcript parses; oversized session truncates to <24k and captures HTTP 200)
• No credentials, API keys, or secrets are included

🤖 Generated with Claude Code

[github-actions]

OB1 PR Gate

✅ Folder structure — All files are in allowed directories
✅ Required files — README.md and metadata.json found in all contribution folders
✅ Metadata valid — All metadata.json files passed JSON Schema validation
✅ No credentials — No API keys, tokens, or secrets detected
✅ SQL safety — No destructive SQL or core table modifications
✅ Category artifacts — Required file types present for each category
✅ PR format — Title follows [category] Description format
✅ No binary blobs — No oversized or binary files
✅ README completeness — All READMEs include Prerequisites, Steps, and Expected Outcome
✅ Contribution dependencies — All declared skill and primitive dependencies exist and are linked in README
✅ LLM clarity review — Covered by Claude PR Review workflow
✅ Remote MCP pattern — No local MCP server patterns detected — uses remote MCP correctly
✅ Tool audit link — Extensions/integrations link to the MCP Tool Audit guide
✅ Scope check — All changes are within the contribution folder(s)
✅ Internal links — All relative links in READMEs resolve to existing files

Result: All 15 checks passed! Ready for human review.

---

Post-Merge Tasks

These don't block merge — they're reminders for admins after this PR lands.

• Add Auto-Capture Claude Code Adapter to skills/README.md
• Add Auto-Capture Claude Code Adapter to root README.md community contributions section
• Add Alan Shurafa (@alanshurafa) to CONTRIBUTORS.md
• Post in OB1 Discord #show-and-tell