Agent skill

Tokf Discover

by mpecan in mpecan/tokf

Find missed token savings in Claude Code sessions and create filters for unfiltered commands

MITAuto-check passed

Install Tokf Discover

skills CLI
$ npx skills add mpecan/tokf --skill tokf-discover -a claude-code

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

GitHub CLI
$ gh skill install mpecan/tokf tokf-discover --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/mpecan/tokf.git skills-src && mkdir -p .claude/skills && cp -r skills-src/crates/tokf-cli/skills/tokf-discover .claude/skills/tokf-discover && 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
tokf-discover
GitHub stars
199
Token cost
~575 tokens
SKILL.md length
253 words
Files
1
Skills in repo
4
Repo updated
First seen
Licence
MIT

At a glance

Find missed token savings in Claude Code sessions and create filters for unfiltered commands

  • Works in 4 steps: Run tokf discover to identify top… → For commands with existing filters: run… → For commands without filters: use… → …
  • SKILL.md covers Quick Start, Options, Interpreting Results and Workflow, plus 2 more sections
  • Calls jq

What it does

Tokf Discover is an agent skill from mpecan/tokf. Find missed token savings in Claude Code sessions and create filters for unfiltered commands

Its SKILL.md is about 580 tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Config-driven CLI tool that compresses command output before it reaches an LLM context. The licence is MIT.

Example prompts

  • “/tokf-discover”

Workflow steps

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

  1. Run tokf discover to identify top savings opportunities
  2. For commands with existing filters: run tokf hook install to set up automatic filtering
  3. For commands without filters: use /tokf-filter skill to create a custom filter
  4. Re-run tokf discover after changes to verify improvement

What it can do on your machine

Read from SKILL.md and the folder at commit 87c93d9. 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:

    • jq

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

  • Network

    No URLs in SKILL.md.

    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

Tokf Discover loads about 575 tokens when it runs. Until then it costs about 27 tokens; SKILL.md has 253 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~27
When it runs · the whole SKILL.md, loaded when a task matches
~575

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 mpecan/tokf at commit 87c93d9, republished under its MIT licence (© mpecan). 253 words, ~575 tokens.

Download SKILL.mdSave it as .claude/skills/tokf-discover/SKILL.md (or your agent's skills folder).
name
tokf-discover
description
Find missed token savings in Claude Code sessions and create filters for unfiltered commands
user_invocable
true

tokf discover — Find Missed Token Savings

Use this skill to analyze Claude Code sessions and find commands that are running without tokf filtering, wasting tokens on verbose output.

Quick Start

Run tokf discover in the project directory to scan recent sessions:

bash
tokf discover

Options

  • --all — scan all projects, not just the current one
  • --since 7d — only scan sessions from the last 7 days (also 24h, 30m)
  • --limit 0 — show all results (default: top 20)
  • --json — output as JSON for programmatic use
  • --session <path> — scan a specific session file
  • --project <path> — scan sessions for a specific project path

Interpreting Results

The output shows:

  • COMMAND — the shell command pattern being run without filtering
  • FILTER — the tokf filter that would handle it
  • RUNS — how many times it appeared in sessions
  • TOKENS — estimated token count of unfiltered output
  • SAVINGS — estimated tokens that filtering would save

Workflow

  1. Run tokf discover to identify top savings opportunities
  2. For commands with existing filters: run tokf hook install to set up automatic filtering
  3. For commands without filters: use /tokf-filter skill to create a custom filter
  4. Re-run tokf discover after changes to verify improvement

Creating Filters for Unfiltered Commands

If tokf discover shows commands with no matching filter, create one:

bash
# See what a filter would look like
tokf which "the-command --args"

# Use the tokf-filter skill to create a proper filter
# /tokf-filter

JSON Output

Use --json for integration with other tools:

bash
tokf discover --json | jq '.results[] | select(.estimated_savings > 1000)'

The JSON schema includes:

  • sessions_scanned — number of JSONL files processed
  • total_commands — all Bash commands found
  • already_filtered — commands already using tokf
  • filterable_commands — commands with available filters
  • no_filter_commands — commands with no matching filter
  • estimated_total_savings — total estimated token savings
  • results[] — per-command breakdown sorted by savings

© mpecan, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in crates/tokf-cli/skills/tokf-discover of mpecan/tokf.

Open the folder on GitHubat commit 87c93d9

Compare with similar skills

Tokf Discover 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.

Tokf Discover compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Tokf Discover this skillmpecan/tokf199—~575Automated safety check: PassMIT
Context Savegarrytan/gstack136k—~9.9kAutomated safety check: NotesMIT
Session Handoffsickn33/agentic-awesome-skills47k—~1.8kAutomated safety check: NotesMIT
Sessionanthropics/claude-for-legal9.6k3 repos~478Automated safety check: PassApache-2.0
Session History Searchslopus/happy24k—~3.1kAutomated safety check: PassMIT
Session Handoffdavila7/claude-code-templates33k2 repos~1.6kAutomated safety check: PassMIT

Similar skills

  • Context Save

    garrytan/gstack

    Captures git state, decisions made and remaining work so that a later session can resume the task without losing context.

    136k GitHub stars~9.9k tokensUpdated today
    Agent WorkflowsAuto-check: notes
  • Session Handoff

    sickn33/agentic-awesome-skills

    A skill your agent uses when context approaches capacity, before /clear or /compact, when switching tasks, or when ending a coding session: produces a structured handoff artifact for the next session.

    47k GitHub stars~1.8k tokensUpdated yesterday
    Agent WorkflowsAuto-check: notes
  • Session

    anthropics/claude-for-legal

    Official

    Run a focused N-question study session on a subject — MBE, essay, or flashcards.

    9.6k GitHub starsUsed in 3 repos~478 tokens
    EducationAuto-check passed
  • Searches past Claude Code, Codex and Cursor sessions and summarizes what was worked on, tried or decided, using extraction scripts instead of reading raw logs.

    24k GitHub stars~3.1k tokensUpdated today
    Agent WorkflowsAuto-check passed
  • Session Handoff

    davila7/claude-code-templates

    Creates comprehensive handoff documents for seamless AI agent session transfers.

    33k GitHub starsUsed in 2 repos~1.6k tokens
    Agent WorkflowsAuto-check passed
  • Read Session

    asgeirtj/system_prompts_leaks

    Locate and read Muse Code's OWN session logs — the current session or a prior one.

    69k GitHub stars~2.4k tokensUpdated yesterday
    Auto-check passed

More from mpecan/tokf

  • Tokf Filter

    mpecan/tokf

    This skill should be used when the user asks to "create a filter", "write a tokf filter", "add a filter for <tool", "how do I filter output", or needs guidance on tokf filter step types, templates…

    199 GitHub stars~9.9k tokensUpdated yesterday
    Auto-check passed
  • Tokf Run

    mpecan/tokf

    Compress verbose CLI output with tokf before returning results.

    199 GitHub stars~571 tokensUpdated yesterday
    Auto-check passed
  • Tokf Discover

    mpecan/tokf

    Find missed token savings by scanning AI coding session files for commands that ran without tokf filtering.

    199 GitHub stars~305 tokensUpdated yesterday
    Auto-check passed

Questions about Tokf Discover

What does Tokf Discover do?

Find missed token savings in Claude Code sessions and create filters for unfiltered commands. Tokf Discover is an agent skill from mpecan/tokf.

How do I install Tokf Discover in Claude Code?

Run `npx skills add mpecan/tokf --skill tokf-discover -a claude-code`. Or copy the skill folder (crates/tokf-cli/skills/tokf-discover in mpecan/tokf) into .claude/skills/tokf-discover in your project. Claude Code loads it when a task matches its description.

How do I install Tokf Discover in Codex?

Run `npx skills add mpecan/tokf --skill tokf-discover -a codex`. Or copy the skill folder (crates/tokf-cli/skills/tokf-discover in mpecan/tokf) into .agents/skills/tokf-discover in your project. Codex loads it when a task matches its description.

Can I use Tokf Discover 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 mpecan/tokf --skill tokf-discover -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/tokf-discover, .gemini/skills/tokf-discover, .github/skills/tokf-discover and .opencode/skills/tokf-discover in your project.

What does Tokf Discover need to run?

Going by SKILL.md and its folder, Tokf Discover needs the command-line tools its instructions call (jq).

Does Tokf Discover access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Tokf Discover 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 Tokf Discover use?

Tokf Discover is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Tokf Discover use?

About 575 tokens (SKILL.md is roughly 2.3k 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 Tokf Discover?

Skills that share tags, products or a category with Tokf Discover: Context Save (garrytan/gstack, 136k stars), Session Handoff (sickn33/agentic-awesome-skills, 47k stars), Session (anthropics/claude-for-legal, 9.6k stars) and Session History Search (slopus/happy, 24k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Tokf Discover?

mpecan (a GitHub user) maintains it in mpecan/tokf, which has 199 GitHub stars. The repository holds 4 skills in this directory. The repository was last updated on October 10, 2026.

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