agentleFS
Sign inStart free
dork-labsdork-labs/dorkos/.claude/skills/reading-session-transcripts/SKILL.md

reading-session-transcripts

Skill11 starsChanged 48 days agoMITView on GitHub

Resolves DorkOS session URLs to JSONL transcript files on disk and reads them. Use when the user shares a session URL, mentions "read this transcript", "read this chat", or references a DorkOS session they want analyzed.

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the Read tool on the resolved path.

    line 58
  2. Distinctive:

    When the user provides multiple URLs, resolve and read each one.

    line 62
  3. Distinctive:

    Session ID without URL: If the user provides just a session ID, ask for the directory (or check the current working directory slug).

    line 69

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Reading Session Transcripts

Overview

DorkOS sessions from the claude-code runtime are backed by Claude SDK JSONL transcript files on disk. When a user shares a session URL or asks you to read a transcript/chat, resolve the URL to the file path and read it directly.

Caveat: session storage is runtime-owned (ADR-0310). This resolution logic only applies to claude-code sessions — sessions from the codex or opencode runtimes live in their own runtime's store, not in ~/.claude/projects/.

When to Use

  • User shares a URL like http://localhost:*/session?session=...&dir=...
  • User says "read this transcript", "read this chat", "look at this session", or similar
  • User pastes one or more session URLs inline with other instructions
  • User references a session ID and directory path

URL Resolution

Given a URL like:

http://localhost:6241/session?session=05da5015-ed83-491c-a07b-10b21379c3e4&dir=%2FUsers%2Fjane%2FMy+Project

Step 1: Extract Parameters

  • Session ID: from session query param → 05da5015-ed83-491c-a07b-10b21379c3e4
  • Directory: from dir query param, URL-decoded → /Users/jane/My Project

Step 2: Slugify the Directory

Replace all non-alphanumeric, non-dash characters with dashes:

/Users/jane/My Project → -Users-jane-My-Project

This matches the SDK's getProjectSlug() in apps/server/src/services/runtimes/claude-code/sessions/transcript-reader.ts.

Step 3: Construct the File Path

~/.claude/projects/{slug}/{sessionId}.jsonl

Full example:

/Users/doriancollier/.claude/projects/-Users-jane-My-Project/05da5015-ed83-491c-a07b-10b21379c3e4.jsonl

Step 4: Read the File

Use the Read tool on the resolved path. The file contains one JSON object per line — the full conversation transcript including messages, tool calls, and results.

Multiple Transcripts

When the user provides multiple URLs, resolve and read each one. They may want comparison, analysis, or context from several sessions.

Edge Cases

  • URL uses a non-standard port: The port doesn't matter — only session and dir query params are needed
  • Spaces in directory: URL-encode as + or %20 — both decode correctly
  • File not found: The session may have been deleted, the directory path may be wrong, or — likely — the session belongs to another runtime (codex or opencode, ADR-0310), whose transcripts don't live in ~/.claude/projects/. Never fail silently: report the resolved path and say the session may belong to a non-claude-code runtime.
  • Session ID without URL: If the user provides just a session ID, ask for the directory (or check the current working directory slug).

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free