feat: runtime monkey-patch instead of file patching — zero fork required

The plugin now monkey-patches HonchoMemoryProvider._do_session_init at
plugin load time instead of requiring changes to any file in the Hermes
repo. This means:

- No modifications to plugins/memory/honcho/__init__.py or any other file
- No fork needed — pure plugin approach
- Survives Hermes upgrades without merge conflicts
- Removed docs/honcho-injector.md (replaced by docs/how-it-works.md)
- Updated README with new architecture description

Config file at /opt/data/identity-config.json remains the single source
of truth for mappings.
This commit is contained in:
2026-05-24 16:07:02 -04:00
parent 172fa90b25
commit 514706bcce
4 changed files with 238 additions and 190 deletions

View File

@@ -1,116 +0,0 @@
# Honcho Injector Patch
The identity plugin cannot directly inject `runtime_user_peer_name` into
Honcho from a user-installed hook (the hook system only allows message
rewrite/skip, not user_id modification). Instead, we add ~6 lines in the
bundled Honcho plugin's session initialization.
## File
`plugins/memory/honcho/__init__.py` — line ~362
## The change
### Before
```python
runtime_user_peer_name=kwargs.get("user_id") or None,
```
### After
```python
runtime_user_peer_name=kwargs.get("user_id") or _resolve_identity_peer(),
```
### Add this function in the same file (or import from the identity plugin)
```python
def _resolve_identity_peer() -> str | None:
"""Resolve peer name from env vars and identity config.
Priority: HERMES_HONCHO_PEER_NAME → kanban task body context_peer
→ kanban board config → None (falls through to Honcho default).
"""
import json, os, sqlite3, re
from pathlib import Path
# 1. Explicit env override
explicit = os.environ.get("HERMES_HONCHO_PEER_NAME")
if explicit:
return explicit
# 2. Kanban worker: read task body for context_peer
task_id = os.environ.get("HERMES_KANBAN_TASK")
db_path = os.environ.get("HERMES_KANBAN_DB")
if task_id and db_path:
db = Path(db_path)
if db.exists():
try:
conn = sqlite3.connect(str(db))
conn.row_factory = sqlite3.Row
try:
row = conn.execute(
"SELECT body FROM tasks WHERE id = ?", (task_id,)
).fetchone()
if row and row["body"]:
m = re.search(r"context_peer:\s*(\S+)", row["body"])
if m:
return m.group(1)
finally:
conn.close()
except Exception:
pass
# 3. Identity config file
cfg_path = Path("/opt/data/identity-config.json")
if cfg_path.exists():
try:
cfg = json.loads(cfg_path.read_text())
# Check kanban board config
board = os.environ.get("HERMES_KANBAN_BOARD")
if board and board in cfg.get("boards", {}):
return cfg["boards"][board]
# Use fallback
return cfg.get("fallback_peer")
except Exception:
pass
return None
```
## How it works
1. When a **gateway session** starts, `kwargs["user_id"]` carries the platform
ID (Discord snowflake, Telegram UID). The identity config file maps these
to canonical peer names. The injector is bypassed — normal flow.
2. When a **kanban worker** starts, `kwargs["user_id"]` is None (no gateway).
The injector kicks in:
a. Checks `HERMES_HONCHO_PEER_NAME` env var (set by a future dispatcher
enhancement or manually).
b. Reads the kanban task body from the SQLite database, extracts
`context_peer: <name>` from the body.
c. Falls back to the board-level config from identity-config.json.
d. If nothing resolves → returns None → Honcho creates `user-default-*`
as today (safe fallback).
## Why not modify the kanban dispatcher?
The kanban dispatcher (`hermes_cli/kanban_db.py:_default_spawn`) is core
Hermes code. We avoid touching it. Instead, the Honcho injector reads the
task directly from the kanban DB using env vars that are already set
(`HERMES_KANBAN_TASK`, `HERMES_KANBAN_DB`, `HERMES_KANBAN_BOARD`).
This adds ~10 microseconds to worker startup — negligible.
## Compatibility
- If the identity plugin is removed, the injector function returns None and
Honcho behaves exactly as before.
- If the config file is missing, same safe fallback.
- No data loss risk.

82
docs/how-it-works.md Normal file
View File

@@ -0,0 +1,82 @@
# Architecture: Why monkey-patch instead of patching files?
The identity plugin uses **runtime monkey-patching** — no files in the
Hermes repo are modified. This is the key design decision that eliminates
the need for a fork.
## The problem
Kanban workers start with NO gateway user identity. The spawn function
runs `hermes -p <profile> chat -q "work kanban task <id>"`. When Honcho
initializes, `kwargs["user_id"]` is None, so it creates `user-default-*`
peers.
There is no Hermes plugin hook that fires at agent initialization before
Honcho creates the session. The hooks available are:
- `pre_gateway_dispatch` — only fires for gateway sessions (Discord, etc.)
- `on_session_start` — fires AFTER Honcho has already initialized
- `pre_tool_call` / `post_tool_call` — fire during tool execution, too late
## The solution: runtime monkey-patch
Python allows replacing any method on any class at runtime. The identity
plugin does exactly this in its `register()` function:
1. At plugin load time, it imports `HonchoMemoryProvider`
2. Stores a reference to the original `_do_session_init` method
3. Replaces it with a wrapper that checks the identity config first
4. If the wrapper finds a peer name, it injects it via `kwargs["user_id"]`
5. Otherwise, it calls the original method unchanged
```python
# In the plugin's register() function:
provider_cls = plugins.memory.honcho.HonchoMemoryProvider
_original_init = provider_cls._do_session_init
@wraps(_original_init)
def wrapper(self, cfg, session_id, **kwargs):
if not kwargs.get("user_id"):
resolved = get_peer_name() # reads config + env vars
if resolved:
kwargs["user_id"] = resolved
return _original_init(self, cfg, session_id, **kwargs)
provider_cls._do_session_init = wrapper
```
## Advantages over file patching
| Aspect | File patching | Monkey-patching |
|--------|--------------|-----------------|
| Repo modifications | Yes (edits plugin file) | Zero |
| Fork needed | Yes (git diff to maintain) | No |
| Upgrades | Merge conflict on every pull | Unchanged |
| Uninstall | Need to revert file changes | Plugin removed → clean |
| Reliability | Works across restarts | Applied at every load |
| Complexity | Simple file edit | Wrapping pattern |
## When does the patch apply?
- At plugin discovery time (Hermes startup)
- Before any Honcho session is created
- Before any agent is initialized
- Before any kanban worker is spawned
The Hermes plugin loader calls `register()` during gateway/agent startup,
which means the patch is in place before any session can begin.
## Safety guarantees
1. **Non-destructive**: The original method is preserved via closure —
removing the plugin restores original behavior with zero cleanup.
2. **Idempotent**: `_apply_honcho_patch()` guards against double-patching
via the `_original_init` sentinel.
3. **Graceful degradation**: If the config file is missing, parsing fails,
or the kanban DB is unavailable, the wrapper returns None → Honcho
falls back to its default behavior (creating `user-default-*` peers).
4. **No data loss**: The injected `kwargs["user_id"]` only affects what
peer name Honcho assigns to the session's messages. It doesn't touch
any existing data.