New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Coding Trace Raw

by uw-syfi in uw-syfi/TraceLab

Read and explain raw Claude Code and Codex CLI session logs in this coding-trace repo.

Apache-2.0Auto-check passedBusiness, Finance & HR

Install Coding Trace Raw

skills CLI
$ npx skills add uw-syfi/TraceLab --skill coding-trace-raw -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install uw-syfi/TraceLab coding-trace-raw --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/uw-syfi/TraceLab.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/coding-trace-raw .claude/skills/coding-trace-raw && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
coding-trace-raw
GitHub stars
138
Token cost
~1.3k tokens
SKILL.md length
553 words
Files
2
Skills in repo
7
Repo updated
First seen
Licence
Apache-2.0

At a glance

Read and explain raw Claude Code and Codex CLI session logs in this coding-trace repo.

  • Works in 5 steps: Work from the coding-trace repo root,… → Identify the provider before… → Prefer the public examples for orientation → …
  • Inspecting examplesessions
  • SKILL.md covers Overview, First Steps, Claude Raw Logs and Codex Raw Logs, plus 2 more sections
  • Calls uv and python

What it does

Coding Trace Raw is an agent skill from uw-syfi/TraceLab. Read and explain raw Claude Code and Codex CLI session logs in this coding-trace repo. Use when inspecting examplesessions, raw .claude/.codex JSONL records, provider record families, duplicate-looking runtime layers, raw tool-call pairing, exposed or omitted fields, token accounting records before normalization, or representative raw windows from findrepresentativesessionsegments.py.

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Business, Finance & HR, covering Accounting and bookkeeping and Database schema design. The repository describes itself as: An open toolkit and public dataset hub for collecting, sanitizing, analyzing, and visualizing coding agent traces. The licence is Apache-2.0.

When your agent uses it

  • Inspecting examplesessions
  • Raw .claude/.codex JSONL records
  • Provider record families
  • Duplicate-looking runtime layers

Example prompts

  • “/coding-trace-raw”

Requirements

  • Python 3

Workflow steps

5 steps, taken from the first numbered list in SKILL.md.

  1. Work from the coding-trace repo root, identified by pyproject.toml, README.md, and scripts/collect_llm_traces.py.
  2. Identify the provider before interpreting records: Claude Code raw logs and Codex CLI raw logs have different schemas.
  3. Prefer the public examples for orientation
  4. Treat public trace.json examples and generated *.expanded.json files as human-readable record dumps with separators. Individual JSON…
  5. Avoid reading large private trace files wholesale. Use targeted sed, rg, head, python -c, or…

What it can do on your machine

Read from SKILL.md and the folder at commit 11b8b14. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • uv
    • python

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use uv, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Coding Trace Raw loads about 1.3k tokens when it runs. Until then it costs about 102 tokens; SKILL.md has 553 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~102
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from uw-syfi/TraceLab at commit 11b8b14, republished under its Apache-2.0 licence (© uw-syfi). 553 words, ~1,314 tokens.

Download SKILL.mdSave it as .claude/skills/coding-trace-raw/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
coding-trace-raw
description
Read and explain raw Claude Code and Codex CLI session logs in this coding-trace repo. Use when inspecting example_sessions, raw .claude/.codex JSONL records, provider record families, duplicate-looking runtime layers, raw tool-call pairing, exposed or omitted fields, token accounting records before normalization, or representative raw windows from find_representative_session_segments.py.

Coding Trace Raw

Overview

Use this skill to reason about raw coding-agent trace files before they are converted into normalized round rows. Keep the distinction clear: raw traces preserve provider-specific runtime records, while normalized traces collapse those records into one JSONL row per LLM invocation.

First Steps

  1. Work from the coding-trace repo root, identified by pyproject.toml, README.md, and scripts/collect_llm_traces.py.
  2. Identify the provider before interpreting records: Claude Code raw logs and Codex CLI raw logs have different schemas.
  3. Prefer the public examples for orientation:
    • example_sessions/claude/explanation.md
    • example_sessions/codex/explanation.md
    • example_sessions/claude/trace.json
    • example_sessions/codex/trace.json
  4. Treat public trace.json examples and generated *.expanded.json files as human-readable record dumps with separators. Individual JSON blocks are valid, but the whole expanded file is not one JSON document.
  5. Avoid reading large private trace files wholesale. Use targeted sed, rg, head, python -c, or scripts/find_representative_session_segments.py.

Claude Raw Logs

Claude Code raw records commonly include:

  • user: visible user messages and tool result messages. Tool results are stored as user records with message.content[] blocks of type tool_result.
  • assistant: model output records. One assistant message can be split across several records that share the same message.id.
  • attachment: auxiliary runtime context such as skill listings or task reminders.
  • system, last-prompt, ai-title, mode, permission-mode, file-history-snapshot: UI/runtime bookkeeping, not separate LLM calls.

When interpreting Claude:

  • Do not count each split assistant record as a separate LLM invocation. Deduplicate by assistant message.id when reading usage.
  • Pair tool calls by matching assistant tool_use.id to later user tool_result.tool_use_id.
  • Read token accounting from message.usage: input_tokens, cache_creation_input_tokens, cache_read_input_tokens, and output_tokens.
  • Treat cache_read_input_tokens as prompt-cache hit prefix tokens.
  • Treat cache_creation_input_tokens + input_tokens as new prompt-side work for that API call.
  • Do not infer readable private thinking from thinking blocks. Public examples preserve that a thinking block existed, not its contents.
Show full SKILL.md (264 more words)Show less

Codex Raw Logs

Codex CLI raw records commonly include:

  • session_meta: session metadata, base instructions, source, model provider, cwd, and git metadata.
  • turn_context: per-turn runtime context such as cwd, sandbox, model, date, and collaboration mode.
  • response_item: structured API-stream items such as messages, reasoning placeholders, function calls, and function-call outputs.
  • event_msg: runtime/UI events such as task_started, user_message, agent_message, token_count, and task_complete.

When interpreting Codex:

  • Expect duplicate-looking layers. A user-visible message can appear as both response_item and event_msg.user_message; an assistant reply can appear as both response_item and event_msg.agent_message.
  • Pair tool calls by response_item.payload.type == "function_call" and later function_call_output records with the same call id.
  • Treat event_msg.token_count as accounting for the preceding model round. last_token_usage.input_tokens already includes cached tokens.
  • Treat cached_input_tokens as prompt-cache hit prefix tokens and input_tokens - cached_input_tokens as uncached prompt-side work.
  • Treat reasoning_output_tokens as a subset of output_tokens, not an additional count.

Finding Examples

Use the finder when the user needs compact raw examples from local histories:

bash
uv run python scripts/find_representative_session_segments.py --provider claude codex --export-dir artifacts/raw_windows

Useful options:

  • --provider claude or --provider codex to focus one format.
  • --min-tool-calls, --min-tool-results, --min-usage, and --min-token-count to require richer windows.
  • --max-file-size-mb 0 only when a full scan is acceptable.
  • --export-dir to write raw JSONL, expanded text, per-record JSON, and a manifest.

Reporting Guidance

When answering raw-trace questions:

  • State whether the evidence is raw-provider behavior or normalized interpretation.
  • Explain duplicate-looking records as runtime layers unless the data proves an extraction bug.
  • Avoid exposing private raw contents in summaries. Prefer counts, schema fields, and short paraphrases unless the user explicitly asks to inspect a public example.
  • If the user wants cross-provider metrics, switch to $coding-trace-normalize or $coding-trace-analyze after identifying the raw source.

© uw-syfi, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/coding-trace-raw of uw-syfi/TraceLab.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit 11b8b14

Compare with similar skills

Coding Trace Raw next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Coding Trace Raw compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Coding Trace Raw this skilluw-syfi/TraceLab138—~1.3kAutomated safety check: PassApache-2.0
Bio Proteomics Data ImportGPTomics/bioSkills1.2k1 repos~4.5kAutomated safety check: PassMIT
Self AwarenessJimLiu/science-skills2272 repos~3.4kAutomated safety check: PassApache-2.0
Sync Upstreamnyaruka/phonenumbers1.6k—~2.8kAutomated safety check: PassMIT
Longbridge Value Investinghelsome/folio2692 repos~1.2kAutomated safety check: PassMIT
Radiology Tablehuang-sir1/radiology-skills1.9k—~1.3kAutomated safety check: PassCustom licence

Similar skills

  • Bio Proteomics Data Import

    GPTomics/bioSkills

    Loads mass-spectrometry data into Python/R and strips the search engine's bookkeeping before any number is trusted -- removes decoys (REV/Reverse), contaminants (CON/Potential contaminant)…

    1.2k GitHub starsUsed in 1 repo~4.5k tokens
    Business, Finance & HRAuto-check passed

  • Self Awareness

    JimLiu/science-skills

    Claude Science's own session database schema and SDK surface for introspection via host.query().

    227 GitHub starsUsed in 2 repos~3.4k tokens
    DatabasesAuto-check passed

  • Sync Upstream

    nyaruka/phonenumbers

    Sync this Go port with a new upstream google/libphonenumber release — regenerate the embedded metadata and reconcile the ported Java logic.

    1.6k GitHub stars~2.8k tokensUpdated 5 days ago
    Business, Finance & HRAuto-check passed

  • Longbridge Value Investing

    helsome/folio

    Value investing analysis using Graham (NCAV/net-net/defensive-investor) and Buffett (economic moat/ROE/FCF) methodologies.

    269 GitHub starsUsed in 2 repos~1.2k tokens
    Business, Finance & HRAuto-check passed

  • Radiology Table

    huang-sir1/radiology-skills

    Create/audit editable publication tables with source reconciliation; not figures or statistical inference.

    1.9k GitHub stars~1.3k tokensUpdated 17 days ago
    Business, Finance & HRAuto-check passed

  • Odoo Agency Fleet Review

    erpipe-org/mcp-odoo

    Review many client Odoo databases at once through odoo-mcp's cross-instance tools — fleet-wide accounting health, per-client aging, partial-failure triage — for agencies and partners managing 5–50…

    420 GitHub stars~699 tokensUpdated 1 mo ago
    Business, Finance & HRAuto-check passed

More from uw-syfi/TraceLab

  • Publish Tracelab Huggingface

    uw-syfi/TraceLab

    Prepare, publish, refresh, or validate the TraceLab public dataset on Hugging Face under UW-SyFI/TraceLab.

    138 GitHub stars~1.4k tokensUpdated 1 mo ago
    Auto-check passed

  • Coding Trace Analyze

    uw-syfi/TraceLab

    Run, summarize, plot, validate, and export normalized coding-trace JSONL files.

    138 GitHub stars~2.5k tokensUpdated 1 mo ago
    Auto-check passed

  • Coding Trace Collect

    uw-syfi/TraceLab

    Collect, count, and extract Claude Code and Codex CLI local histories into normalized coding-trace JSONL files.

    138 GitHub stars~1.9k tokensUpdated 1 mo ago
    Auto-check: notes

  • Coding Trace Normalize

    uw-syfi/TraceLab

    Explain and work with the normalized coding-trace JSONL row format produced by extractclauderounds.py, extractcodexrounds.py, and collectllmtraces.py.

    138 GitHub stars~1.6k tokensUpdated 1 mo ago
    Auto-check passed

  • Coding Trace Sanitize

    uw-syfi/TraceLab

    Sanitize normalized coding-trace JSONL rows for public sharing.

    138 GitHub stars~1.3k tokensUpdated 1 mo ago
    Auto-check: notes

  • Tracelab Public Release

    uw-syfi/TraceLab

    Publish TraceLab snapshots from the internal repo to the public uw-syfi/TraceLab GitHub repo as clean, mergeable, incremental pull requests from a persistent public mirror branch.

    138 GitHub stars~2.7k tokensUpdated 1 mo ago
    Auto-check passed

Categories

Questions about Coding Trace Raw

What does Coding Trace Raw do?

Read and explain raw Claude Code and Codex CLI session logs in this coding-trace repo. Coding Trace Raw is an agent skill from uw-syfi/TraceLab. Read and explain raw Claude Code and Codex CLI session logs in this coding-trace repo.

When should I use Coding Trace Raw?

Coding Trace Raw fits situations like: inspecting examplesessions; raw .claude/.codex JSONL records; provider record families; duplicate-looking runtime layers.

How do I install Coding Trace Raw in Claude Code?

Run `npx skills add uw-syfi/TraceLab --skill coding-trace-raw -a claude-code`. Or copy the skill folder (skills/coding-trace-raw in uw-syfi/TraceLab) into .claude/skills/coding-trace-raw in your project. Claude Code loads it when a task matches its description.

How do I install Coding Trace Raw in Codex?

Run `npx skills add uw-syfi/TraceLab --skill coding-trace-raw -a codex`. Or copy the skill folder (skills/coding-trace-raw in uw-syfi/TraceLab) into .agents/skills/coding-trace-raw in your project. Codex loads it when a task matches its description.

Can I use Coding Trace Raw in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add uw-syfi/TraceLab --skill coding-trace-raw -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/coding-trace-raw, .gemini/skills/coding-trace-raw, .github/skills/coding-trace-raw and .opencode/skills/coding-trace-raw in your project.

What does Coding Trace Raw need to run?

Going by SKILL.md and its folder, Coding Trace Raw needs the command-line tools its instructions call (uv and python). Our summary lists: Python 3.

Does Coding Trace Raw access the network?

SKILL.md contains no URLs. Its commands use uv, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Coding Trace Raw safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Coding Trace Raw use?

Coding Trace Raw is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Coding Trace Raw use?

About 1.3k tokens (SKILL.md is roughly 5.3k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Coding Trace Raw?

Skills that share tags, products or a category with Coding Trace Raw: Bio Proteomics Data Import (GPTomics/bioSkills, 1.2k stars), Self Awareness (JimLiu/science-skills, 227 stars), Sync Upstream (nyaruka/phonenumbers, 1.6k stars) and Longbridge Value Investing (helsome/folio, 269 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Coding Trace Raw?

uw-syfi (a GitHub organization) maintains it in uw-syfi/TraceLab, which has 138 GitHub stars. The repository holds 7 skills in this directory. The repository was last updated on August 22, 2026.

Source: uw-syfi/TraceLab on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.