All plugins

approval-ledger

Local 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 network

olric3015Toolsv1.2.0Updated 1 month ago

About

Local 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 network

approval-ledger

A local ledger of the approval prompts Hermes raised, and an /approval-log command that shows
which rules ask most, how the prompts ended and how long you took to answer.

Hermes asks before it runs a dangerous command (a recursive delete, sudo, a force push, ...),
before a protected write, and for MCP consent. Each prompt is answered, denied, times out or is
withdrawn, and then it is gone. approval-ledger subscribes to pre_approval_request and
post_approval_response and keeps one row per decision, so you can see afterwards that one rule
asked thirty times this week, that most prompts on your phone time out, or that smart mode
denied something you then had to approve by hand.

Install

hermes plugins install approval-ledger --enable

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

Use

/approval-log             last 24 hours, by the rule that asked
/approval-log session     the most recent session that asked
/approval-log all         everything in the ledger
/approval-log 7d          the last 7 days (any number of hours or days: 6h, 30d)
/approval-log surfaces    add to any of the above to group by where it asked
/approval-log days        add to any of the above to group by day, newest first
/approval-log recent      add to any of the above to list the newest 10 decisions instead
/approval-log recent 25   the newest 25 instead (1 to 100)
/approval-log clear       delete the ledger

(/approvals is Hermes's own command for the approval mode; this plugin only reads.)

Example (/approval-log all, from the end-to-end test on Hermes 0.21.5: two prompts answered by
hand, then smart mode denying one command, the person declining to override it, and smart mode
approving another):

Approvals, all recorded:

pattern           asked  approved  denied  unanswered  median wait
recursive delete      5         2       3           0         0.6s

Smart-mode verdicts count as approved or denied; median wait is people's answers only.
  • asked counts decisions. A smart-mode deny that Hermes then offers you to override is two.
  • approved is once, session or always (or a smart-mode approve); denied is deny (or a
    smart-mode deny).
  • unanswered is a prompt nobody answered: it timed out, the turn ended or was interrupted
    before an answer, or the notification could not be delivered. The command did not run.
  • median wait is how long a person took to answer, from the prompt appearing to the answer.
    Smart-mode verdicts and unanswered prompts are left out of it.
  • pattern is the label of the rule that asked (recursive delete, sudo, ...), never the
    command. Grouped by surfaces, the first column is where it asked instead: cli, gateway
    (messaging platforms, the TUI and Desktop), smart, transport:<plugin>, or an MCP/vault
    consent surface.

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

Approvals, all recorded, newest 5 of 5:

when             pattern  surface         answer  wait
1s ago  recursive delete    smart  smart_approve  0.0s
1s ago  recursive delete      cli           deny  0.6s
2s ago  recursive delete    smart     smart_deny  1.0s
4s ago  recursive delete      cli           once  0.3s
4s ago  recursive delete      cli           deny  0.6s

/approval-log days puts one day on each line, newest day first, so a day of many prompts or
many unanswered ones stands out. Days are your computer's local calendar days, and a day with no
decisions has no line. Example (the plugin's own output for a sample week):

Approvals, last 7 days:

day         asked  approved  denied  unanswered  median wait
2026-10-08      3         2       1           0         4.2s
2026-10-07     12         6       2           4         3.4s
2026-10-06      2         1       1           0         6.6s
total          17         9       4           4         3.9s

Smart-mode verdicts count as approved or denied; median wait is people's answers only.

A time window counts back from now (/approval-log 6h, /approval-log 7d surfaces,
/approval-log 30d days, /approval-log 30d recent). 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

  • Commands Hermes never asks about: an approval you already gave for the session or
    permanently, approvals.mode: off, yolo mode, a sandboxed backend that skips the check, and
    one-shot (-q) or cron runs, where nobody could answer, so Hermes blocks or approves without
    a prompt.
  • A smart-mode verdict of "escalate" is not a decision: the prompt that follows it is.

What is recorded

One JSON line per decision in <HERMES_HOME>/plugin-data/approval-ledger/approvals.jsonl (per
profile):

ts, session_id, surface, pattern, choice, outcome, decided_by, coalesced,
wait_s.

Not recorded: the command, its description, the session key (it can hold a platform chat id),
the deny reason. The hooks hand the plugin the command and description; the plugin drops both.
The session key is held in memory only, for as long as the prompt is open, to pair the answer
with its prompt.

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

Security and footprint

  • register() only registers two hooks (pre_approval_request, post_approval_response) and
    one command (/approval-log).
  • Observer only: it cannot answer or change an approval. Each hook call notes a time or appends
    one line to the ledger, and nothing else.
  • No network access, no subprocesses, no downloads, no credentials read.
  • Writes only approvals.jsonl under its own plugin data directory. /approval-log clear
    deletes it.
  • The hooks never raise: if the ledger cannot be written, the failure is logged at debug level
    and the approval flow is not affected.
  • Standard library only.

Development

python -m pytest tests                                                # unit tests, no Hermes needed
PYTHONPATH=<hermes checkout> python tests/e2e/approval_ledger_e2e.py   # Hermes's own approval guard
hermes plugins validate approval-ledger

Changes

  • 1.2.0: /approval-log recent takes a row count, /approval-log recent 25 (1 to 100;
    ten without one). What is recorded did not change.
  • 1.1.0: /approval-log days, one line per day with decisions, newest day first; works
    with every window (/approval-log 30d days). What is recorded did not change.
  • 1.0.0: first release.

MIT licensed.

README from the reviewed commit 4a2ef86; it updates when the author re-pins.

More by olric3015

18 plugins by olric3015 in the catalog.

olric3015aux-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.Communitytoolsolric3015command-ledgerLocal ledger of the slash commands you run in Hermes (built-in commands in the CLI, built-in and plugin commands on messaging platforms): /command-log shows how often each command was used, its share, in how many sessions and when last, per command, platform, surface or day, for the last 24 hours, any number of hours or days, the latest session or everything recorded; /command-log recent lists the newest commands (ten, or up to 100 with /command-log recent 50) with the alias typed. Disclosure — writes command metadata (command name, alias typed, surface, platform, a 12-digit hash of the session key; no arguments, session key or user id) to plugin-data/command-ledger/commands.jsonl, trimmed to 5,000 rows past 2 MB; observer only, cannot block or change a command; no networkCommunitytools

Developer Mode

Explore built-in Hermes toolsets and actions.

HermesAgent

Developer?

Create a Hermes extension for Desktop or agent workflows.

Developer Section