Codigest

Share the code context you choose with an AI: your project, a few files, what changed, or the command that failed.

한국어

_{Automated release-specific security scan (A–F), not an independent security certification. How ratings work.}

Codigest is a command-line tool for the moment you ask ChatGPT, Claude or a coding agent about your code. It turns the part of your project you pick into one block of text you can paste into any chat.

A model only knows what you show it. Copying files by hand is slow, it is easy to miss the one file that matters, and it is just as easy to paste far more than the question needs. Codigest lets you decide what to share and then collects it for you: the whole project, a file and the files it imports, what changed since a point you marked, or a failing test with the code it points at.

What it does

• Bundles code you pick. The whole project, chosen files and folders, or the local files they import (Python, JS/TS), as one block with the file tree.
• Shows what changed. Text or function-level diffs since a baseline you set, without touching your Git history.
• Brings in terminal output. It records commands you run and can hand over a failing one together with the files its error points at.
• Says what it included. Every bundle reports its size, and pack gives the reason for each file and names the ones it left out.
• Keeps secrets out. .env files and keys are never included, and tokens in code or output are masked.
• Works with agents. Claude Code, Cursor and other MCP clients can request the same output.

Codigest does not call any AI model; it only prepares what you send to one. Everything runs on your machine.

_{One of the workflows below: a test fails, and one command collects what the model needs.}

Install

Prebuilt binary, no Python needed. Windows (PowerShell):

irm https://raw.githubusercontent.com/NaNIsaac/codigest/main/install.ps1 | iex

Linux, macOS and Termux:

curl -fsSL https://raw.githubusercontent.com/NaNIsaac/codigest/main/install.sh | sh

This installs cdg to %LOCALAPPDATA%\Programs\codigest (added to your user PATH) on Windows, ~/.local/bin on Linux and macOS, and $PREFIX/bin in Termux. Run the same line again to update. CDG_VERSION=0.12.1 pins a version and CDG_INSTALL_DIR=... picks the folder (then PATH is left alone). Supported: Windows x64, Linux x64 and ARM64, macOS Intel and Apple Silicon, Android ARM64 (Termux; the tests also run inside Termux in CI). Archives for download by hand are on the releases page. No setup is needed in your projects. Both scripts check the archive against the SHA-256 GitHub records for it. On Linux and macOS, install.sh will not replace a cdg from the Python package in the same folder unless CDG_FORCE=1 is set.

Python package

The same commands are on PyPI as codigest. Install it with uv, which also fetches the Python version it needs (3.15) and keeps it apart from your projects:

winget install --id astral-sh.uv                  # Windows (skip if you have uv)
curl -LsSf https://astral.sh/uv/install.sh | sh   # macOS/Linux

uv tool install codigest
uv tool update-shell                              # then restart your terminal

The command is codigest, or cdg for short. Both installs provide a command called cdg, so keep only one of them on PATH. If the install fails with a syntax error on lazy import, run uv self update and try again.

Next, cdg term install hooks your shell so the output of every command is kept for cdg pack and cdg term. It is off until you run it; see Terminal capture.

On Windows, PowerShell's help codigest is a separate command from codigest --help. For optional native Get-Help integration without automatic command capture:

cdg shell-help install       # register comment-based PowerShell help
# open a new PowerShell window
help codigest               # or: Get-Help codigest
cdg shell-help uninstall     # remove it when no longer needed

This adds a small forwarding function to your PowerShell profile; other help commands are unchanged. The CLI's own full command reference is still available via cdg --help.

In a terminal, every command below copies its result to the clipboard. Add -s to also print it, -o FILE to also write it to a file, --no-copy to leave the clipboard alone, and -m "..." to put your question or instruction in it. When the output goes to a pipe or a file instead (cdg scan > ctx.xml, cdg digest | less, a script, a coding agent's shell), it is printed and the clipboard is left alone; add --copy to copy it anyway. Over SSH the copy goes to the clipboard of the machine you type on (OSC 52), not the remote one; see output options.

1. Share your project with an AI chat

Run scan in your project and paste:

cdg scan

$ cdg scan
Copied snapshot to clipboard
Snapshot 3 files, ~121 tokens

The clipboard now holds the file tree and every file, honoring .gitignore:

<project_snapshot project="project">
<project_structure>
├── .gitignore
├── prices.py
└── test_prices.py
</project_structure>
<source_code>
<file path=".gitignore">
.venv/
__pycache__/
.pytest_cache/
</file>

<file path="prices.py">
def apply_discount(price, percent):
    return price * (1 - percent)
</file>

<file path="test_prices.py">
from prices import apply_discount


def test_ten_percent():
    assert apply_discount(200, 10) == 180
</file>
</source_code>
</project_snapshot>

Large projects ask before copying (above 30k tokens or 100 files). For those, share less, as in the next section.

2. Share only part of it, or what changed

cdg scan src/app.py -r       # one file and the local files it imports, recursively
cdg scan src/billing/ docs/  # folders, files, or both
cdg scan src/app.py -l       # with line numbers, so answers can say "line 42"
cdg scan src/app.py:40-80    # only lines 40 to 80, numbered (src/app.py:42 for one line)
cdg digest                   # signatures and docstrings only, much smaller
cdg digest src/billing/      # the same for one folder
cdg tree -L 2                # the file tree, two levels deep

The examples from here on are the same project a little later: prices.py has grown a Cart class, and an old MATLAB version sits in legacy/.

digest is the overview to start a conversation with. It lists every function, class and top-level variable of Python files, and the functions and %% sections of MATLAB files, in the order they appear, with the line they start on and the first line of their docstring (or, in MATLAB, of their help comment). Class attributes and dataclass fields are listed with their values, shortened when long:

$ cdg digest -s
<code_digest project="project" contents="signatures and docstrings, no bodies">
<project_structure>
├── .gitignore
├── legacy
│   └── discount.m
├── prices.py
└── test_prices.py
</project_structure>
<file path="legacy/discount.m">
1: function p = discount(price, percent)
       Price after taking off percent percent.
3: %% Apply
</file>
<file path="prices.py">
 1: TAX_RATE = 0.1
 4: def apply_discount(price: float, percent: float) -> float
        Price after taking off `percent` percent.
 9: class Cart
        Items and the discount that applies to all of them.
12:   def __init__(self, percent: float=0)
16:   def total(self) -> float
          Sum after discount and tax.
</file>
<file path="test_prices.py">
4: def test_ten_percent()
</file>
</code_digest>
Digest 3 files, ~192 tokens

A model that has seen this can ask for exactly what it needs ("show me Cart.total"), and cdg scan prices.py:16-19 gives it just that method, numbered the same way:

$ cdg scan prices.py:16-19 -s
<project_snapshot project="project">
<project_structure>
└── prices.py
</project_structure>
<source_code>
<file path="prices.py" lines="16-19">
16:     def total(self) -> float:
17:         """Sum after discount and tax."""
18:         subtotal = sum(apply_discount(p, self.percent) for p in self.items)
19:         return subtotal * (1 + TAX_RATE)
</file>
</source_code>
</project_snapshot>

tree -L keeps a large or vendored folder from flooding the output:

$ cdg tree -L 1
├── .gitignore
├── legacy/ (1 file)
├── prices.py
└── test_prices.py

To share what changed while you work, set a baseline first:

cdg mark                     # e.g. before a refactor
# ... work ...
cdg diff                     # text diff since the mark
cdg semdiff                  # Python: which functions and classes changed

A baseline is a private copy kept in .codigest/; your Git history is not touched.

3. Hand over a failing command

Run the command through cdg cap. The output looks exactly as it would without it; Codigest keeps a copy. Then ask for a pack:

cdg cap -- pytest -q        # or npm test, cargo test, go test ./..., make
cdg pack

$ cdg pack
Packed ~370 / 30,000 tokens
  failed   pytest -q (exit 1, just now)
  changes  uncommitted changes (git)
  + test_prices.py  line 5 in failing output
  + prices.py       changed
Copied pack to clipboard

Read it as: the failing command, your uncommitted Git changes, test_prices.py because the error points at its line 5, and prices.py because you edited it.

<context_pack project="project">
<terminal_history>
<command index="1" command="pytest -q" exit_code="1" cwd="~/project" duration="1.5s" at="2026-10-17T01:58:14+09:00">
<output>
F                                                                      [100%]
================================= FAILURES ==================================
_____________________________ test_ten_percent ______________________________

    def test_ten_percent():
>       assert apply_discount(200, 10) == 180
E       assert -1800 == 180
E        +  where -1800 = apply_discount(200, 10)

test_prices.py:5: AssertionError
========================== short test summary info ==========================
FAILED test_prices.py::test_ten_percent - assert -1800 == 180
1 failed in 0.11s
</output>
</command>
</terminal_history>
<changes source="uncommitted changes (git)">
diff --git a/prices.py b/prices.py
index 3d691df..428752a 100644
--- a/prices.py
+++ b/prices.py
@@ -1,2 +1,2 @@
 def apply_discount(price, percent):
-    return price * (1 - percent / 100)
+    return price * (1 - percent)
</changes>
<file path="test_prices.py" reason="line 5 in failing output">
1: from prices import apply_discount
2: 
3: 
4: def test_ten_percent():
5:     assert apply_discount(200, 10) == 180
</file>
<file path="prices.py" reason="changed">
1: def apply_discount(price, percent):
2:     return price * (1 - percent)
</file>
</context_pack>

How pack decides

Plain cdg pack fills its token budget (default 30,000) in this order:

1. Files named in the output of the last failed command run in this project in the past hour: File "x.py", line 7 (Python), x.py:7 (pytest, ruff, gcc, rustc, eslint), x.ts(7,3) (tsc). Once the same command passes again, that failure is no longer used. A file that is only named in the output, not changed by you, comes as an excerpt: ten lines around each place the output points at, marked with a lines="40-60" attribute. Short files come whole.
2. The project functions and classes that the failing Python tests call, whole, with reason called from failing test .... A traceback that ends in an assertion names only the test; this brings the code under test even when the bug is already committed and nothing shows in a diff.
3. Files you changed: uncommitted and untracked files in Git, or changes since cdg mark. A long changed file comes as the parts its diff hunks touch (the diff itself is in the pack already).
4. Tests that import those files (test_*.py, *_test.py, *.test.ts, *.spec.js, ...).
5. Project files they import: Python imports, and relative imports in JS/TS (./x, ./x.js for x.ts, ./dir for dir/index.ts).

When you name files, you get only those files. --with adds automatic sources back:

cdg pack src/prices.py                         # just this file
cdg pack src/prices.py --with tests            # this file and its tests
cdg pack --with changes                        # only your changes, no failure
cdg pack --with failure,changes,tests,imports  # the same as plain `cdg pack`

The failed command and the diff are always included when their source is on. A long command output keeps its first 80 and last 40 lines (the first failures and the summary); cdg term -f shows all of it. Tokens are estimated at four characters each, so the budget is approximate.

Making it smoother

cdg term install hooks your shell so every command is captured without typing cdg cap (cdg term uninstall removes it). cdg term shows what was captured; see Terminal capture.

4. Let an agent fetch it

Connect your coding agent over MCP and it requests context itself instead of asking you to paste:

claude mcp add --scope user codigest -- cdg mcp      # Claude Code

Other clients (Cursor, VS Code, Claude Desktop): add a stdio server with command cdg and args ["mcp"]. The tools are listed under MCP tools.

An agent that can run shell commands needs no setup at all. Its shell is not a terminal, so cdg prints the result for it and never touches your clipboard. A way of working that suits agents:

cdg tree -L 2                      # where things are
cdg digest src/                    # what is defined where, with line numbers
cdg scan src/prices.py:16-19      # read the parts that matter, numbered the same way
cdg mark                           # before editing
cdg semdiff                        # afterwards: which functions it changed
cdg cap -- pytest -q && cdg pack   # run the tests; on failure, the pack has what to fix

The code in every block is exactly as written in the file (->, <, & are not escaped), so code copied out of it can be pasted back. A large scan with nobody to answer the size question stops with a hint instead of waiting; -y takes it all.

Reference

• All commands
• Terminal capture
• MCP tools
• Configuration
• Privacy and safety
• Upgrading from 0.9
• How it works
• Development

All commands

cdg --help lists them in the same groups.

Shared options: -s / --stdout also prints the result (status messages go to stderr; on by default when piped), --copy / --no-copy overrides the clipboard default (copy in a terminal, not when piped), -o FILE also writes it to a file, -m / --message adds an instruction. scan -l adds line numbers, scan -a ignores the config filters, scan -y skips the confirmation (CI). -r / --resolve on scan, diff, semdiff and digest also pulls in local files imported by the targets. scan and term take --format markdown|json for chats that render Markdown or for scripts (XML is the default). --preset NAME on any command fills in options saved under [presets.<command>.NAME] in .codigest/config.toml or your user config; options typed after it win. cdg <command> --help lists everything.

Terminal capture

cdg cap streams the command's output live and records the command line, combined stdout/stderr, exit code and working directory. cdg term --context also includes the hostname at capture time and the shell name. The PC name is opt-in because it may be sensitive; older captures have no hostname, and cdg term still copies without it. Captures from all projects share one store on your machine, so cap and term work in any directory.

cdg cap "npm run build | tail -20"  # quote pipelines
cdg term              # last command and its output
cdg term -n 3         # last 3
cdg term -f           # last failed one
cdg term --context   # add the PC hostname and shell to the copied context (opt-in)
cdg term --here -n 3  # last 3 run in this project
cdg term -i 2         # only the command from 2 runs ago
cdg term -n 3 --format json  # the last 3 as JSON (markdown also works)
cdg term path         # where captures are stored (override with CODIGEST_HOME)
cdg term clear        # delete all captures

Commands not run through cap are read from the native shell history (PowerShell, bash, zsh, fish), without their output.

Automatic capture

cdg term install edits your PowerShell profile, ~/.bashrc and ~/.zshrc for whichever of these shells is installed; choose with --shell powershell|bash|zsh|all. On Windows, Git Bash counts as bash. Commands still run in your current session (cd, variables and functions work normally) and appear in your history as typed. Programs that need a real terminal are skipped: editors, pagers, ssh, full-screen tools (tmux, zellij, herdr, k9s, yazi, ...), docker/podman/kubectl with -it, npm/pnpm/yarn/bun init or create, npx create-*, pip uninstall, gh auth login, az login, gcloud auth login and gcloud init, aws configure, ssh-keygen, sudo, and database shells (psql, mysql, sqlite3, ...). A prompt that does not end its line stays invisible in the capture pipe until the next newline, so the command would look stuck.

• Skip one command: start the line with a space.
• Skip more programs: $env:CODIGEST_CAP_SKIP = '^\s*(myrepl|mytui)\b' (bash/zsh: export CODIGEST_CAP_SKIP='^(myrepl|mytui)( |$)', an extended regex).
• Disable for a session: $env:CODIGEST_NOCAP = 1 (bash/zsh: export CODIGEST_NOCAP=1).
• Remove the hooks: cdg term uninstall.

The profile loads a copy of the hook that cdg term install puts in the data folder (shell/ next to the captures), not the file inside the package. After updating codigest, run cdg term install again to refresh it.

Limitations:

• Captured programs write to a pipe, not a TTY, so some lose colors or progress bars.
• In bash (4+) and zsh, multi-line commands (loops typed over several lines, heredocs), background jobs (cmd &) and lines using ! history expansion are not captured; they run normally. In bash, neither are lines that start with a subshell ( ... ).
• In bash the hook uses the DEBUG trap and keeps one set before the first prompt (starship, bash-preexec); a DEBUG trap set later replaces it.
• Long output is bounded: the PowerShell hook keeps the first and last 2,000 lines, and a bash or zsh capture file over 4 MB is read 1 MB from each end. The store then keeps 64 KB.
• In PowerShell, a command counts as failed when its exit code is non-zero or an error reached the screen; an error hidden with -ErrorAction is not a failure.
• The PowerShell hook does not wrap lines in remote sessions, nested prompts or the debugger.

MCP tools

The tools return the same text the CLI prints. None of them moves the baseline, and paths outside the project the server was started in are refused.

Configuration

cdg init writes .codigest/config.toml. The [filter] section controls which files every command considers:

[filter]
# Only these extensions (omit to include every non-ignored file)
extensions = [".py", ".ts", ".rs", ".md", ".json"]

# Extra exclusions, in .gitignore syntax (.gitignore itself is always honored)
exclude_patterns = [
    "*.lock",
    "tests/data/",
    "legacy_code/"
]

.git/, .codigest/, virtual environments, node_modules/ and __pycache__/ are always skipped.

Privacy and safety

• Credential files. .env* (except .env.example and similar templates), *.pem, *.key, *.p12, *.pfx, keystores, .netrc, .pgpass and SSH private keys are never included by any command. scan says which it withheld, pack lists them as omitted, and their hunks in a diff are replaced with a notice.
• Tokens in code and diffs. Private keys, JWTs, vendor tokens (OpenAI, GitHub, AWS, Google, Slack, GitLab, npm, Stripe sk_/rk_, Hugging Face hf_) and passwords in URLs are masked in everything Codigest outputs. Ordinary code such as token = get_token() is left alone.
• Secrets in captures. On top of that, terminal output is checked before it is written to disk for Authorization headers and for key=value and key: value pairs, with a quoted key ("password": "...") or a key that ends in a secret name after _ or - (AWS_SECRET_ACCESS_KEY, DB_PASS; also pass and pwd). The output of credential-printing commands is never stored: env, printenv, gh auth token, aws configure and aws sts get-session-token/assume-role, az account get-access-token, gcloud auth print-access-token, kubectl get secret, kubectl config view --raw, op read, bw get, pass show, vault read, cat of .env, .npmrc, .pypirc, .netrc, key files or credentials.json, and a few more (the full list is DENY_COMMANDS in src/codigest/core/redact.py).
• These rules catch common formats, not every possible secret.
• Your project. No command changes your source files. Only mark and init create .codigest/, which holds its own .gitignore (*) so it never shows in git status; init also adds it to the project's .gitignore.
• Where captures live. cdg term path shows the file. On macOS and Linux it is readable only by you. cdg term clear deletes it.
• Nothing leaves your machine. Codigest makes no network requests. Context goes where you paste it, or to the MCP client you connected.

Upgrading from 0.9

• Piped or redirected output is printed without -s and no longer copied; add --copy if a script relied on the clipboard.
• <file> blocks hold the code verbatim instead of XML-escaped. A custom template in .codigest/prompts.toml now receives the blocks unescaped too; only {instruction} is escaped.
• digest output is numbered, in source order, with docstrings, and takes several paths.

Upgrading from 0.8

• scan, diff, semdiff and digest no longer save into .codigest/*.txt; use -o FILE. diff --no-save and the like are gone.
• cdg init no longer sets a baseline; run cdg mark.
• cdg pack FILES packs only those files; add --with failure,changes,tests,imports for the previous behavior.
• Outputs no longer start with built-in instructions to the model ("You are an expert..."). Put your own with -m, or a template in .codigest/prompts.toml.

How it works

• Baseline. A Git repository at .codigest/anchor/.shadow_git, separate from your project's repository. mark commits the current state to it; diff, semdiff, status and pack compare the working tree against it.
• Captures. One newline-delimited JSON file holds the last 200 commands; each output is capped at 64 KB, keeping the head and tail.
• Output. File contents go into <file> blocks verbatim. Only a closing tag of one of Codigest's own blocks inside a file (a literal </file>) is written as </file>, so it cannot end the block early.
• MCP. The server is a small JSON-RPC loop over stdio. Each tool runs the regular CLI with -s --no-copy, so an agent sees exactly what you would paste.

Development

git clone https://github.com/NaNIsaac/codigest
cd codigest
uv run cdg --help              # run from the checkout
uv run --with pytest pytest -q # tests

License

MIT. See LICENSE.