ashlr-plugin

Install with npm, call the CLI with --json, and register the ashlr-mcp server.

ashlr-plugin is an open source efficiency layer for coding agents. It gives Claude Code, Codex, Cursor, and other MCP hosts shorter file, search, edit, and shell results, then records the difference in a local ledger.

Agents spend their context on tool output. Shorter results leave more room for the task. On the 2026-10-17 reference run, reads were about 75% smaller and greps about 27% smaller per call. That is not a whole-session total. The same run's per-repo means were about 60% on vercel/ai, about 63% on pandas, and about 55% on tokio. Reproduce it with bun run scripts/run-benchmark.ts --compare. Methodology.

The local plugin is MIT licensed. No account required. Telemetry is off unless you opt in (ASHLR_TELEMETRY=on): opt-in telemetry only. It ships 54 MCP tools and the full slash-command catalog.

Agent index: llms.txt. Runnable reference: llms-full.txt. Rendered docs: llms.txt and llms-full.txt.

This illustrates one read mode. By default, large supported JavaScript and TypeScript files can instead return an AST skeleton; other reads may return compacted or summarized content. mode: "snip" selects head-and-tail compaction. See the tool catalog.

Install

Requires Bun 1.3 or newer. No account required. One command installs the CLI and the MCP server:

npm install -g [email protected]

Confirm the install, list the tools, and read the local ledger. Each command prints JSON on stdout:

ashlr-plugin --ashlr-handshake
ashlr-plugin tools --json
ashlr-plugin stats --json
ashlr-plugin codex-doctor --json
ashlr-plugin codex-install --dry-run --json

ashlr-plugin version prints the version as text. ashlr-plugin mcp starts the stdio router. From a checkout of this repo, the same commands are bun run scripts/cli.ts <subcommand>. The handshake is answered by the ashlr-plugin program, not by cli.ts.

Claude Code plugin (slash commands, redirects, and the status line). This clones the plugin into the Claude marketplace cache. It does not install the global npm bin:

curl -fsSL https://plugin.ashlr.ai/install.sh | bash

Then, inside Claude Code:

/plugin marketplace add ashlrai/ashlr-plugin
/plugin install ashlr@ashlr-marketplace
/reload-plugins

Codex plugin:

codex plugin marketplace add ashlrai/ashlr-plugin
codex plugin add ashlr@ashlr-marketplace
ashlr-plugin codex-doctor --json

Windows (PowerShell):

irm https://raw.githubusercontent.com/ashlrai/ashlr-plugin/main/docs/install.ps1 | iex

MCP server

After the npm install, ashlr-mcp is on PATH. Paste one snippet. Do not edit a path.

Claude Code, project file .mcp.json:

{
  "mcpServers": {
    "ashlr": {
      "command": "ashlr-mcp",
      "args": [],
      "env": {
        "ASHLR_MCP_HOST": "claude-code"
      }
    }
  }
}

Cursor, project file .cursor/mcp.json:

{
  "mcpServers": {
    "ashlr": {
      "command": "ashlr-mcp",
      "args": [],
      "env": {
        "ASHLR_MCP_HOST": "cursor"
      }
    }
  }
}

Codex, add this block to ~/.codex/config.toml. It matches ashlr-plugin codex-install --dry-run --json:

[mcp_servers.ashlr]
command = "ashlr-mcp"
args = []
cwd = "."
env = { ASHLR_MCP_HOST = "codex-cli", ASHLR_CODEX_HOOK_MODE = "nudge" }

Goose and other hosts: ports/README.md and docs/multi-host-mcp.md.

How it works

1. The host starts ashlr-mcp over stdio.
2. The agent calls ashlr__read, ashlr__grep, ashlr__edit, or ashlr__bash instead of the noisy built-in.
3. The tool performs its operation and returns a compact result. File and search tools operate within the allowed project roots; ashlr__grep requires ripgrep (rg) on PATH.
4. Recorded savings are written to ~/.ashlr/stats.json. Read them with ashlr-plugin stats --json or ashlr__savings.

Claude Code can also redirect built-in tools when ASHLR_HOOK_MODE=redirect. Codex nudges and does not force a redirect. Cursor calls the MCP tools directly.

Tool calls

Arguments below match the tool catalog. Run them only after the MCP server is connected. ashlr__read and ashlr__grep do not modify files.

Read this README:

{
  "name": "ashlr__read",
  "arguments": {
    "path": "README.md"
  }
}

Search the server source:

{
  "name": "ashlr__grep",
  "arguments": {
    "pattern": "registerTool",
    "cwd": "servers"
  }
}

Read the ledger:

{
  "name": "ashlr__savings",
  "arguments": {}
}

ashlr__edit takes path, search, and replace. It writes the file. Use it when you mean to change that file. The full catalog is site/content/docs/tools and, on the web, plugin.ashlr.ai/docs/tools.

Features

Need	Call
Read a large file	ashlr__read
Search	ashlr__grep
Edit without echoing the file	ashlr__edit
Shell output	ashlr__bash
Long-running shell commands	ashlr__bash_start, ashlr__bash_tail, ashlr__bash_stop, ashlr__bash_list
Repo map	ashlr__tree or ashlr__orient
Tests and diffs	ashlr__test or ashlr__diff_semantic
GitHub PRs and issues	ashlr__pr or ashlr__issue
Savings	ashlr__savings or ashlr-plugin stats --json

54 MCP tools are registered by default. One experimental tool, ashlr__pipe, stays off until ASHLR_PIPE_ENABLE=1.

FAQ

Does an agent need an account?

No account required. npm install -g and the MCP snippet are enough for local tools and the ledger.

Where do file contents go?

The MCP server runs on the machine that started it. Tool results go back to the host, the same way a built-in read would. Telemetry is off unless ASHLR_TELEMETRY=on. The contract is docs/telemetry.md.

What does a savings number mean?

It is the difference for one tool call, not the cost of a whole session. The published reference run is docs/benchmarks.md. Your own ledger is ashlr-plugin stats --json.

Which hosts get slash commands?

Claude Code gets slash commands, hook redirects, and the status line. Codex gets the plugin, skills, nudge hooks, and the CLI. Cursor gets the MCP server. The matrix is docs/multi-host-mcp.md.

What is Pro?

The local plugin is the free product. Pro ($12/month or $120/year) adds cloud summarization and cross-machine stats. Team ($24/user/month, or $20/user/month billed annually, minimum 3) adds a shared encrypted genome. Prices and limits: docs/pricing.md.

Star

If this saved context on a real task, star the repo so other people and agents can find it: github.com/ashlrai/ashlr-plugin.

Releasing

Automatic releases

Merge a version bump (package.json) to main on green. auto-tag.yml waits for every required check on that exact commit, creates v<version> with the ashlr-release App token, and release.yml tests and publishes it. No designated caller and no manual step.

Manual, one command each:

gh workflow run auto-tag.yml -R ashlrai/ashlr-plugin
gh workflow run release.yml -R ashlrai/ashlr-plugin -f version=X.Y.Z

The first tags the current main version if its tag is missing and checks are green. The second builds and publishes an existing vX.Y.Z tag.

Release notes

Version history lives in CHANGELOG.md.

---

Status-line

The Claude Code status bar shows savings and other session information.

Install:

bun run scripts/install-status-line.ts

With the sparkline and tips disabled, examples include:

Example output	Meaning
ashlr · ctx: 50% · session +0 ≈$0.00 · lifetime +1.5K (est.)	No session savings yet; the lifetime total is estimated.
ashlr · ctx: 92% · session +0 ≈$0.00	Context usage is available from Claude Code input; the lifetime segment is absent when its selected total is zero.

Fields are separated by ·; + denotes saved tokens. Counts use K for thousands and M for millions. Optional fields depend on settings, available data and terminal width.

Field	Meaning
ashlr	Identifies the status line.
Heartbeat and 7d sparkline	The heartbeat animates briefly after a save when animation is enabled. Seven cells show daily tokens saved, oldest to newest, for the current and preceding six UTC days, scaled to the largest day. The 7d label appears above 100 columns.
ctx: 50%	Context used as a percentage of the limit supplied by Claude Code; hidden when usable input is unavailable.
session ↑+12.3K	Tokens saved across matching current-session buckets. ↑ (or ASCII +) marks a save in the last four seconds; without stats the line may say waiting for first tool call, or collecting… during early zero-savings use.
≈$0.01	Estimated session savings in US dollars, using the detected summarizer's input price, an ASHLR_PRICING_MODEL override, or main-model pricing when the detected price is zero and no override is set. Values below one cent display as ≈$0.00.
lifetime +1.2M (est.)	Local lifetime tokens saved. After ten measured calls, (API-measured) instead shows only the measured-call total, which can be lower than the overall estimate.
* 3d streak	Stored current streak of active days; shown from three days.
☁ 2 machines	Machine count from the cached cross-machine aggregate for Pro; the cache can be stale.
brief: concise	Active user-level brief setting; hidden when off.
last: ashlr__read +5.0K [2s ago]	Age of the latest session save and the highest-saving tool entry in a matching session bucket. The count is that entry's accumulated savings, so the tool and count are a proxy, not necessarily the last call. Shown for up to 60 seconds.
tip: … or ↑: …	A rotating usage tip, or an upgrade hint when an apparently free user has saved at least 50,000 session tokens.

On narrow terminals, fields can be omitted or the line truncated; an absent field does not imply a zero value. See the status-line documentation for visibility rules, fallback behavior and stale data.

The optional statusLine settings live under the ashlr key in ~/.claude/settings.json. Omitted booleans use the defaults shown here.

Field	Default	Effect
statusLine	true	Master switch; false hides the status line.
statusLineSession	true	Show the current session's saved-token count and cost estimate.
statusLineLifetime	true	Show lifetime saved-token count when it is greater than zero.
statusLineTips	true	Add a rotating tip, or an eligible upgrade nudge, when it fits.
statusLineSparkline	true	Show the 7-day activity sparkline and heartbeat.
statusLineUpgradeNudge	true	Allow the free-tier 50K-session-savings upgrade nudge; set false to silence it.

For example, to hide tips and lifetime totals while keeping session savings:

{
  "ashlr": {
    "statusLineTips": false,
    "statusLineLifetime": false
  }
}

The status line is Claude Code-specific. Codex users can inspect savings with ashlr-plugin stats --json or the ashlr__savings MCP tool.

---

Provider base URLs

The plugin's own provider calls (the summarizer and token-count calls to Anthropic, the genome reachability probes for Anthropic and OpenAI, and the Ollama checks in /ashlr-ollama-setup and genome-init) read the usual environment variables and keep their old defaults when none is set:

Variable	Default	Form
ANTHROPIC_BASE_URL	https://api.anthropic.com	root, without /v1
OPENAI_BASE_URL	https://api.openai.com/v1	includes /v1
OLLAMA_HOST	http://localhost:11434	host, host:port, :port or a full URL

Trailing slashes are trimmed. Pointing these at a gateway such as a LiteLLM proxy or a Phantom proxy is therefore possible, because the plugin just sends its requests to whatever URL you set. Only the URL handling is covered by tests here; the plugin has not been tested against any specific gateway, and what a gateway does with your keys and requests is documented by that gateway. Not covered: the Ollama probe in the session greeting and the optional ASHLR_EMBED_URL embedding endpoint, which keep their own settings.

Under a Phantom session these variables are set for you; see Plugs into Phantom Secrets.

---

Plugs into Phantom Secrets

Phantom Secrets swaps the real keys in a project for scoped phm_ placeholders and swaps them back at an authenticated local proxy: the agent can use the key but never sees it. The plugin runs inside that boundary with no extra wiring. phantom exec hands the child phm_ placeholders and points ANTHROPIC_BASE_URL and OPENAI_BASE_URL at http://127.0.0.1:<port>/<route>/_phantom/<token>/; the plugin sends whatever key and base URL it is given, so the placeholder flows through unchanged and the proxy does the swap.

# Hosts run inside the Phantom boundary
phantom exec -- claude
phantom exec -- codex

# Plugin scripts work the same way
phantom exec -- bun run scripts/doctor.ts
phantom exec -- bun run scripts/run-benchmark.ts --validate-tokenizer

Inside a session:

• /ashlr-doctor reports it. A phantom exec line shows the proxy port, which provider keys hold phm_ placeholders, and which base URLs point at the session's proxy (session token redacted). It warns when a phm_ placeholder sits outside a session, when a placeholder's base URL does not point at the proxy, or when a key is a phm: reference. The doctor also lists whether the phantom CLI is installed and its version; it runs only phantom --version and never reads or reveals secret values.
• Optional preflight. ASHLR_PHANTOM_PREFLIGHT=1 (or strict) makes the doctor also run Phantom's value-free phantom agent doctor --json and report its verdict as a phantom preflight line. See docs/phantom-preflight.md.
• Benchmarks adapt. run-benchmark --validate-tokenizer skips its API sample when the key would not work from that process, and says why.
• Hard guarantees. Detection uses variable names, the phm_ / phm: prefixes, and the base URL host and port only. The plugin never resolves phm: references, never runs phantom reveal, and never prints key values.

The Phantom fleet mod in mods/ goes further on Claude Code 2.1.309+ with a pinned Locus binding and compatible Phantom/Locus CLIs: governed model routing (phm router select), a /top fleet pane, /peers, and person-acknowledged /send. /ashlr-fleet explains it inside a session. It ships from the git plugin only (the npm package omits mods/), and some of its CLI contracts are still proposals upstream: mods/README.md marks what is verified.

The full phantom exec contract, including LiteLLM notes and the strict-gateway variables, lives in docs/PHANTOM-USAGE.md. Session detection and the doctor line are covered by __tests__/phantom-exec.test.ts and __tests__/doctor-phantom.test.ts; the plugin has not been tested against a live Phantom session here.

---

Commands

Use ashlr-plugin in a terminal and slash commands inside Claude Code. MCP hosts call tools through ashlr-mcp.

Task	Terminal	Claude Code
Start here	ashlr-plugin codex-doctor for Codex diagnostics	/ashlr-start
Check savings	ashlr-plugin stats --json	/ashlr-savings or /ashlr-dashboard
Inspect tools	ashlr-plugin tools	/ashlr-help
Preview Codex setup	ashlr-plugin codex-install --dry-run	See multi-host setup
Refresh changed genome files	ashlr-plugin genome-refresh	to inspect the automatic loop

See the complete command reference for every shipped slash command, terminal command, aliases, and source links. Claude Code slash commands are not terminal subcommands. The CLI executable is ashlr-plugin; ashlr belongs to Ashlr Hub.

Free vs Pro

The free tier is the product: 54 MCP tools, the full slash-command catalog, the full genome scribe loop, and every benchmark included. No account required.

Pro ($12/mo, 7-day trial) adds cloud infrastructure for developers who need it: cross-machine stats sync, hosted embedding retrieval, cloud LLM summarizer (no local Ollama required), and a live auto-updating savings badge. Team ($24/user/mo, min 3) adds shared encrypted team genome, org savings dashboard, policy packs, and SSO.

Run /ashlr-upgrade to upgrade, or see plugin.ashlr.ai/docs/pro/pricing for the full comparison.

Architecture

See plugin.ashlr.ai/docs/contributing/architecture for how the tools, hooks, and genome scribe loop fit together. Internal notes: docs/architecture.md.

Changelog

See CHANGELOG.md for version history.

Part of the Ashlr ecosystem

The Ashlr plugin is one of the open-source tools from Ashlr, a forward-deployed engineering company. Its siblings: Lexicon, Locus, Phantom Secrets, Ashlr Stack, Ashlr Markdown.

Contributing

Issues and PRs are welcome. Look for good first issue to get started, and read CONTRIBUTING.md. Report security problems privately per SECURITY.md.

License

MIT: LICENSE.