LispM Agent

Ask, and it gets done.

Talk to a LispM agent in a chat laid out like Claude Code's. Underneath, the conversation is an org document that is also the model's context, its tools are Elisp functions, its approvals are TODO entries, its memory is org files under git, and it evaluates Lisp in the Emacs you are using.

It works in your project.

LispM's agent reads the code, writes the fix and runs the tests, in a chat that is an org document, with tools that are Lisp.

Every agent, one sidebar.

Claude Code, Codex and Hermes work beside LispM's agents, every session listed in the sidebar, to resume where it ran.

And they talk to each other.

Any of them asks another for a review and gets the answer as its next message, and uses LispM's skills and tools as LispM's own agents do.

LispM's classic-style desktop with Claude Code and Codex side by side, both running. Claude Code asks Codex to review a function; Codex gets the request as its next message, and its reply comes back to Claude Code as one.

Characters with a soul of their own.

In the Characters app, give an agent a name, a pixel-art portrait, a soul, a role and a model of its own. Each character—like @Ada, @Milo and @Juno—becomes its own application in the Applications folder that you, Claude Code and Codex can message by name.

LispM's Applications folder showing Ada, Milo and Juno as applications with pixel-art portraits alongside Ada's chat window reviewing a diff.

An 8-room pixel office for every agent.

Press C-c O to open LispM Office: an animated pixel office where every agent session has a desk. Working agents type at their laptops, idle agents hold coffee, waiting agents raise a hand with an approval badge, and clicking any desk opens its inspector, mailbox and pending tool calls.

LispM Office in Focus Mode showing the QA Office pixel room in the middle and the LispM Office Inspector on the right with Work, Mailbox and Character tabs.

LispM itself, one field away.

M-x ai, or the magnifying glass in the menu bar, drops one field under the menu bar for the whole computer while your keyboard stays in the window you were in. As you type, LispM finds apps, control panels, skills and tools; anything else goes to LispM, a standing agent taught by the desktop and its apps that keeps one conversation across restarts.

FrontierHarness 1.0 Benchmark

Tied for #1 pass rate. #1 fastest harness overall.

On the 30 official containerized tasks of the FrontierHarness 1.0 benchmark (Terminal-Bench 2.0 and DeepSWE), LispM ties for #1 in pass rate (66.7%, 20/30 tasks) while finishing successful tasks in a median of 3m 24s (203.9 seconds)—nearly twice as fast as the next fastest harness—on the Pareto efficiency frontier.

Pass Rate (30 Official Tasks)

  1. LispM ★66.7% (20/30)
  2. Codex66.7% (20/30)
  3. DSH (Creator)63.3% (19/30)
  4. Claude Code63.3% (19/30)
  5. Pi60.0% (18/30)
  6. DSH (Standard)60.0% (18/30)
  7. DSH (PTC)60.0% (18/30)
  8. Kimi Code56.7% (17/30)
  9. DSH (Minimal)56.7% (17/30)
  10. Oh My Pi56.7% (17/30)
  11. Exo Harness53.3% (16/30)
  12. Hermes50.0% (15/30)
  13. OpenCode50.0% (15/30)

Median Time per Pass (Speed, lower is faster)

  1. LispM ★3m 24s
  2. DSH (Minimal)5m 41s
  3. DSH (Standard)6m 17s
  4. Exo Harness6m 17s
  5. OpenCode6m 27s
  6. Codex6m 43s
  7. DSH (Creator)6m 44s
  8. Oh My Pi6m 46s
  9. Hermes6m 58s
  10. Pi7m 33s
  11. DSH (PTC)7m 44s
  12. Kimi Code7m 56s
  13. Claude Code9m 38s
#HarnessPass RateMedian Cost / PassEffective Cost / PassWarm Cache HitMedian Time
1LispM ★66.7% (20/30)$0.0809$1.9883.2%3m 24s (203.9s)
2Codex66.7% (20/30)$0.1243$3.4788.0%6m 43s (403.3s)
3DSH (Creator)63.3% (19/30)$0.1194$3.2884.3%6m 44s (403.9s)
4Claude Code63.3% (19/30)$0.2880$18.3467.8%9m 38s (578.0s)
5Pi60.0% (18/30)$0.0709$2.4379.4%7m 33s (453.3s)
6DSH (Standard)60.0% (18/30)$0.1201$3.4686.5%6m 17s (376.5s)
7DSH (PTC)60.0% (18/30)$0.1370$4.5887.2%7m 44s (463.8s)
8Kimi Code56.7% (17/30)$0.1818$3.6588.0%7m 56s (475.9s)
9DSH (Minimal)56.7% (17/30)$0.1214$4.7284.6%5m 41s (341.2s)
10Oh My Pi56.7% (17/30)$0.1354$4.7582.2%6m 46s (406.4s)
11Exo Harness53.3% (16/30)$0.0748$1.0570.3%6m 17s (377.2s)
12Hermes50.0% (15/30)$0.1746$2.9085.9%6m 58s (417.8s)
13OpenCode50.0% (15/30)$0.0615$3.2478.4%6m 27s (387.3s)