All plugins

stop-ledger

Local ledger of the turns you stopped while Hermes was still working (/stop, or /new and /reset during a running turn, on messaging platforms, the TUI and the desktop app): /stop-log shows how many running turns were stopped, how many by /stop and by /new, in how many sessions and how long ago the last one was, per platform or per day, for the last 24 hours, any number of hours or days, the latest session or everything recorded; /stop-log recent lists the newest stops (ten, or up to 100 with /stop-log recent 50). Disclosure — writes stop metadata (platform, /stop or /new, Hermes's label for the stop, a 12-digit hash of the session key; no session key or chat content) to plugin-data/stop-ledger/stops.jsonl, trimmed to 5,000 rows past 2 MB; observer only, cannot delay or undo a stop; no network

olric3015Toolsv1.0.0Updated 1 month ago

About

Local ledger of the turns you stopped while Hermes was still working (/stop, or /new and /reset during a running turn, on messaging platforms, the TUI and the desktop app): /stop-log shows how many running turns were stopped, how many by /stop and by /new, in how many sessions and how long ago the last one was, per platform or per day, for the last 24 hours, any number of hours or days, the latest session or everything recorded; /stop-log recent lists the newest stops (ten, or up to 100 with /stop-log recent 50). Disclosure — writes stop metadata (platform, /stop or /new, Hermes's label for the stop, a 12-digit hash of the session key; no session key or chat content) to plugin-data/stop-ledger/stops.jsonl, trimmed to 5,000 rows past 2 MB; observer only, cannot delay or undo a stop; no network

stop-ledger

A local ledger of the turns you stopped while Hermes was still working, and a /stop-log
command that shows how often that happens, on which platform, in how many sessions and how
recently.

Stopping a turn halfway usually means something went wrong: the agent went off in the wrong
direction, took too long, or was about to do something you did not want. stop-ledger subscribes
to agent_loop_stopped, which Hermes fires when a running turn is interrupted, and keeps one row
per stop. You can see afterwards that you stopped Hermes six times on Telegram today, that most
of those were /new rather than /stop, or that the desktop app gets stopped far more than
your phone.

Install

hermes plugins install stop-ledger --enable

Requires Hermes 0.21.5 or newer. No API key, no configuration, no dependencies.

Use

/stop-log             last 24 hours, by platform
/stop-log session     the most recent session you stopped
/stop-log all         everything in the ledger
/stop-log 7d          the last 7 days (any number of hours or days: 6h, 30d)
/stop-log days        add to any of the above to group by day, newest first
/stop-log recent      add to any of the above to list the newest 10 stops instead
/stop-log recent 25   the newest 25 instead (1 to 100)
/stop-log clear       delete the ledger

Example (/stop-log all, from the end-to-end test: on Telegram a running turn stopped with
/stop and another cleared with /new, and one turn stopped from the TUI):

Stopped turns, all recorded:

platform  stops  by /stop  by /new  sessions    last
telegram      2         1        1         1  3s ago
tui           1         1        0         1  3s ago
total         3         2        1         2  3s ago

Only turns that were still running count: a /stop or /new between turns is not a stop.
  • platform is where the turn ran: a messaging platform (telegram, discord, ...) or
    tui for the TUI and the desktop app.
  • stops counts stopped turns. by /stop is a /stop (or the Stop button in the TUI or
    the desktop app); by /new is a /new or /reset that cleared a turn still running.
  • sessions is how many different sessions those stops were in, and last how long ago
    the newest one was. The platform with the most stops comes first.

/stop-log recent lists the newest ten stops of the chosen window, newest first
(/stop-log recent 25 lists 25, up to 100). From the same run:

Stopped turns, all recorded, newest 3 of 3:

when    platform  cause             detail       session
3s ago       tui   stop  session_interrupt  504e634656fb
3s ago  telegram    new        new_command  5b8bb83e3f6b
3s ago  telegram   stop       stop_command  5b8bb83e3f6b

detail is Hermes's own label for the stop (stop_command, new_command,
session_interrupt, ...), and session a short hash that tells sessions apart without
naming them.

/stop-log days puts one day on each line, newest day first; days are your computer's local
calendar days. A time window counts back from now (/stop-log 6h, /stop-log 7d days). The
ledger keeps its newest 5,000 rows, so a long window on a busy install reaches back only as far
as those rows do.

What it does not see

  • The classic CLI: Hermes has no interrupt hook there (Ctrl+C in the CLI is not recorded).
  • A /stop when no turn is running yet, and a /new between turns: nothing was interrupted.
  • In the TUI and the desktop app, Hermes reports every interrupt of a running turn the same way,
    including one it makes itself (for example when a disconnected client's session is cleaned
    up), so those are counted as stops too.

What is recorded

One JSON line per stopped turn in <HERMES_HOME>/plugin-data/stop-ledger/stops.jsonl (per
profile):

ts, session, platform, cause, detail.

session is the first 12 hex digits of a SHA-256 of the session key, not the key itself: a
gateway session key names the chat and the user. Nothing said in the chat reaches the hook.

The file is bounded: once it passes 2 MB it is rewritten to its newest 5,000 rows.

Security and footprint

  • register() only registers one hook (agent_loop_stopped) and one command (/stop-log).
  • Observer only: it cannot delay or undo a stop. Each hook call appends one line to the ledger,
    and nothing else.
  • No network access, no subprocesses, no downloads, no credentials read.
  • Writes only stops.jsonl under its own plugin data directory. /stop-log clear deletes it.
  • The hook never raises: if the ledger cannot be written, the failure is logged at debug level
    and the stop goes ahead.
  • Standard library only.

Development

python -m pytest tests                                            # unit tests, no Hermes needed
PYTHONPATH=<hermes checkout> python tests/e2e/stop_ledger_e2e.py   # Hermes's own interrupt code
hermes plugins validate stop-ledger

Changes

  • 1.0.0: first release.

MIT licensed.

README from the reviewed commit 604940a; it updates when the author re-pins.

More by olric3015

18 plugins by olric3015 in the catalog.

olric3015approval-ledgerLocal ledger of Hermes's approval prompts (dangerous commands, protected writes, MCP consent, smart-mode verdicts): /approval-log shows how often each rule asked, how many prompts were approved, denied or left unanswered, and how long a person took to answer, per rule, per surface or per day, for the last 24 hours, any number of hours or days, the latest session or everything recorded; /approval-log recent lists the newest decisions (ten, or up to 100 with /approval-log recent 50). Disclosure — writes decision metadata (session id, surface, rule label, answer, timings; no command, description or session key) to plugin-data/approval-ledger/approvals.jsonl, trimmed to 5,000 rows past 2 MB; observer only, cannot answer or change an approval; no networkCommunitytoolsolric3015aux-ledgerLocal ledger of Hermes's auxiliary LLM calls (titling, compression, vision, approval): /aux shows tokens, time and failures per task, model, provider or day, or lists the newest calls (ten, or up to 100 with /aux recent 50), for the last 24 hours, any number of hours or days, the latest session or everything recorded. Disclosure — writes auxiliary-call metadata (ids, models, timings, token counts, no prompt or response text) to plugin-data/aux-ledger/calls.jsonl, trimmed to 5,000 rows past 2 MB; no networkCommunitytoolsolric3015b64-ref@b64: context references that decode encoded text for the model: @b64:<value> takes base64 (standard or URL-safe, padded or not), hex (0x48656c6c6f, 48:65:6c, \x48\x65), percent-encoding or a data: URI, told apart by the value's shape or picked with base64:, base64url:, hex: or url:, and attaches what it holds: text as text with control and invisible characters escaped, JSON pretty-printed, gzip and zlib inflated in memory (at most 1 MiB, one layer), and binary named by its format (PNG and GIF with their size, JPEG, PDF, ZIP, ELF, SQLite and more) with a hex dump of its first 128 bytes; notes point out a value that was cut short, padding that is off, a second layer of encoding and a JWT inside; the TUI and Desktop composer preview a value as you type it. Disclosure — reads only the text of the reference itself, using Python's base64, json, urllib.parse and zlib; nothing decoded is run, opened or saved; no network, no files read or written, no subprocess, no credentials read; the plugin keeps no copy of what it decodes, but the @b64: reference and the decoded contents become part of the message sent to your model provider and saved in the session history, so decode only values you are willing to share with it; register() only adds the @b64: reference provider.Communitytoolsolric3015chmod-ref@chmod: context references that explain a file permission mode for the model: @chmod:<mode> takes an octal mode (755, 0755, 4755), ls -l text (-rwsr-xr-x, with the file type letter and a trailing + . or @) or an st_mode as stat and git write it (100644), and attaches both forms, each class's rights in words (directory rights named as such), what setuid, setgid and the sticky bit do and where they are ignored, and notes such as a mode everyone can change, one private to the owner, or a class holding rights the class before it lacks; @chmod:"644 u+x,go-w" applies a symbolic change step by step the way GNU chmod does (u g o a, + - =, r w x X s t, copies like u=g), @chmod:"100644 100755" shows what changes between two modes, and @chmod:"umask 027" shows what new files and directories get; the TUI and Desktop composer preview a mode as you type it. Disclosure — reads only the text of the reference itself, using Python's re; it never looks at a file on disk; no network, no files read or written, no permission changed, no subprocess, no credentials read; the plugin keeps no copy of what you write, but the @chmod: reference and the explanation become part of the message sent to your model provider and saved in the session history; register() only adds the @chmod: reference provider.Communitytools

Developer Mode

Explore built-in Hermes toolsets and actions.

HermesAgent

Developer?

Create a Hermes extension for Desktop or agent workflows.

Developer Section