Files
hermes-agent/tools/memory_tool.py

417 lines
18 KiB
Python

#!/usr/bin/env python3
"""Memory Tool - persistent curated memory (MEMORY.md = agent notes, USER.md =
user profile). Both enter the system prompt as a FROZEN snapshot at session
start; mid-session writes hit disk immediately but never change the prompt
(prefix cache stays intact). Single `memory` tool: add/replace/remove or a
batch `operations` list. The store lives in ``tools.memory_tool_store``."""
import copy
import json
import logging
from contextvars import ContextVar
from pathlib import Path
from hermes_constants import get_hermes_home
from typing import Dict, Any, List, Optional, Tuple
from utils import is_truthy_value
from tools.registry import no_cache_check_fn
# fcntl is Unix-only; on Windows use msvcrt for file locking. MemoryStore reads
# these lazily from this module (tests inspect ``memory_tool.fcntl``).
msvcrt = None
try:
import fcntl
except ImportError:
fcntl = None
try:
import msvcrt
except ImportError:
pass
logger = logging.getLogger(__name__)
# One tool-definition pass must use one config decision for both availability
# and the dynamic target schema. ContextVar keeps concurrent profile/session
# builds isolated while letting the check_fn result flow to the immediately
# following dynamic_schema_overrides call in ToolRegistry.get_definitions().
_memory_surface_flags: ContextVar[Optional[Tuple[bool, bool]]] = ContextVar(
"memory_surface_flags", default=None
)
def get_memory_dir() -> Path:
"""Return the profile-scoped memories directory (resolved per call so
HERMES_HOME/profile switches after import are respected)."""
return get_hermes_home() / "memories"
from tools.memory_tool_store import ( # noqa: E402,F401 (re-exports)
ENTRY_DELIMITER, MEMORY_BLOCK_HEADERS, MemoryStore, _READ_FAILED,
_drift_error, _read_failed_error, _scan_memory_content,
)
def load_on_disk_store() -> "MemoryStore":
"""Fresh on-disk MemoryStore with configured limits/flags, for contexts with
no live agent (gateway, Desktop, bare CLI ``/memory``) so approvals enforce
the SAME caps as ``agent_init``. Defaults if config can't load; never raises."""
kwargs: Dict[str, Any] = {}
try:
from hermes_cli.config import load_config
config = load_config() or {}
mem_cfg = get_builtin_memory_config(config)
memory_enabled, user_profile_enabled = get_builtin_memory_store_flags(config)
kwargs = {
"memory_char_limit": int(mem_cfg.get("memory_char_limit", 2200)),
"user_char_limit": int(mem_cfg.get("user_char_limit", 1375)),
"memory_enabled": memory_enabled,
"user_profile_enabled": user_profile_enabled,
}
except Exception:
kwargs = {} # config optional — fall back to defaults rather than break /memory
store = MemoryStore(**kwargs)
store.load_from_disk()
return store
# ---------------------------------------------------------------------------
# Write-approval gate
# ---------------------------------------------------------------------------
def _target_label(target: str) -> str:
return "user profile" if target == "user" else "memory"
def _gate_or_stage(summary: str, detail: str, payload: Dict[str, Any]) -> Optional[str]:
"""Run the memory write gate. Returns a JSON tool-result string when the
write must NOT proceed (blocked, or staged for approval), None to proceed.
If the gate module can't load, fail open rather than block all writes."""
try:
from tools import write_approval as wa
except Exception:
return None
decision = wa.evaluate_gate(wa.MEMORY, inline_summary=summary, inline_detail=detail)
if decision.allow:
return None
if decision.blocked:
return tool_error(decision.message, success=False)
record = wa.stage_write(wa.MEMORY, payload, summary=f"{summary}: {detail[:120]}", origin=wa.current_origin())
return json.dumps({"success": True, "staged": True, "pending_id": record["id"], "message": decision.message},
ensure_ascii=False)
def _apply_write_gate(action: str, target: str, content: Optional[str], old_text: Optional[str]) -> Optional[str]:
"""Gate a single mutating op (add/replace/remove); other actions pass."""
if action not in _STORE_ACTIONS:
return None
label = _target_label(target)
if action == "add":
summary, detail = f"add to {label}", content or ""
elif action == "replace":
summary, detail = f"replace in {label}", f"old: {old_text}\nnew: {content}"
else:
summary, detail = f"remove from {label}", old_text or ""
payload = {"action": action, "target": target, "content": content, "old_text": old_text}
return _gate_or_stage(summary, detail, payload)
def _apply_batch_write_gate(target: str, operations: List[Dict[str, Any]]) -> Optional[str]:
"""Gate a whole batch as a single unit."""
summary = f"apply {len(operations)} op(s) to {_target_label(target)}"
detail_lines = []
for op in operations:
op = op or {}
act = op.get("action", "?")
_op_content = op.get("content") or op.get("new_text") or ""
if act == "remove":
detail_lines.append(f"- remove: {op.get('old_text', '')}")
elif act == "replace":
detail_lines.append(f"- replace: {op.get('old_text', '')} -> {_op_content}")
else:
detail_lines.append(f"- {act}: {_op_content}")
payload = {"action": "batch", "target": target, "operations": operations}
return _gate_or_stage(summary, "\n".join(detail_lines), payload)
# ---------------------------------------------------------------------------
# Tool entry point
# ---------------------------------------------------------------------------
def _missing_old_text_error(store: "MemoryStore", target: str, action: str) -> str:
"""Recoverable error for replace/remove without ``old_text``. It can't be
schema-required (needs a combinator the Codex backend rejects — see
test_memory_tool_schema.py) and some clients omit it, so return the current
inventory plus a retry instruction instead of a dead-end."""
return json.dumps({
"success": False,
"error": (f"'{action}' needs old_text -- a short unique substring of the entry "
f"to {action}. None was provided. Reissue the {action} with old_text "
f"set to part of one of the current_entries below."),
"current_entries": store._entries_for(target),
"usage": store._usage(target),
}, ensure_ascii=False)
def _validate_single_op(store, action, target, content, old_text) -> Optional[str]:
"""Validate required params BEFORE the gate so an invalid write is rejected
now rather than staged and failing at approve time."""
if action == "add" and not content:
return tool_error("Content is required for 'add' action.", success=False)
if action in ("replace", "remove") and not old_text:
return _missing_old_text_error(store, target, action)
if action == "replace" and not content:
return tool_error("content is required for 'replace' action.", success=False)
return None
# action -> store call for both the live tool path and staged-write replay.
_STORE_ACTIONS = {
"add": lambda store, target, content, old_text: store.add(target, content),
"replace": lambda store, target, content, old_text: store.replace(target, old_text, content),
"remove": lambda store, target, content, old_text: store.remove(target, old_text),
}
def memory_tool(
action: str = None,
target: str = "memory",
content: str = None,
old_text: str = None,
new_text: str = None,
operations: Optional[List[Dict[str, Any]]] = None,
store: Optional[MemoryStore] = None,
) -> str:
"""Tool entry point; returns a JSON string. Single op (action + content /
old_text) or batch (``operations`` applied atomically against the final
budget). ``new_text`` aliases ``content`` — callers mirror ``old_text``
with it (patch-tool shape), which used to leave ``content`` empty."""
if store is None:
return tool_error("Memory is not available. It may be disabled in config or this environment.", success=False)
if content is None and new_text is not None:
content = new_text
# Strict providers send JSON null for optional fields; treat as omitted.
if target is None:
target = "memory"
target_error = _memory_target_error(store, target)
if target_error is not None:
return json.dumps(target_error)
if operations:
if not isinstance(operations, list):
return tool_error("operations must be a list of {action, content?, old_text?} objects.", success=False)
gate_result = _apply_batch_write_gate(target, operations)
if gate_result is not None:
return gate_result
return json.dumps(store.apply_batch(target, operations), ensure_ascii=False)
run = _STORE_ACTIONS.get(action)
if run is None:
return tool_error(f"Unknown action '{action}'. Use: add, replace, remove", success=False)
invalid = _validate_single_op(store, action, target, content, old_text)
if invalid is not None:
return invalid
# Approval gate: when on, stages the write (background/gateway) or prompts
# inline (interactive CLI); when off (default) passes straight through.
gate_result = _apply_write_gate(action, target, content, old_text)
if gate_result is not None:
return gate_result
return json.dumps(run(store, target, content, old_text), ensure_ascii=False)
def get_builtin_memory_config(config: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
"""Normalized ``memory`` config section ({} when missing/malformed → flags
default to enabled). ``agent_init`` consumes the same section so tool
availability and store construction cannot diverge."""
if config is None:
try:
from hermes_cli.config import load_config_readonly
config = load_config_readonly()
except Exception:
logger.debug("Could not read memory config for availability", exc_info=True)
return {}
section = config.get("memory") if isinstance(config, dict) else None
return section if isinstance(section, dict) else {}
def get_builtin_memory_store_flags(config: Optional[Dict[str, Any]] = None) -> Tuple[bool, bool]:
"""Return ``(memory_enabled, user_profile_enabled)`` from resolved config."""
section = get_builtin_memory_config(config)
return (
is_truthy_value(section.get("memory_enabled"), default=True),
is_truthy_value(section.get("user_profile_enabled"), default=True),
)
@no_cache_check_fn
def check_memory_requirements() -> bool:
"""Snapshot store flags and report whether the built-in tool is available."""
_memory_surface_flags.set(None)
flags = get_builtin_memory_store_flags()
_memory_surface_flags.set(flags)
return flags[0] or flags[1]
def _memory_target_error(store: "MemoryStore", target: str) -> Optional[Dict[str, Any]]:
"""Return a shared validation error for an invalid or disabled target."""
if target not in {"memory", "user"}:
from tools.registry import _bound_error_text
return {"success": False,
"error": _bound_error_text(f"Invalid memory target '{target}'. Use 'memory' or 'user'.")}
if store.target_enabled(target):
return None
label = "USER.md" if target == "user" else "MEMORY.md"
return {"success": False, "error": f"Built-in {label} writes are disabled in memory config.", "target": target}
def apply_memory_pending(payload: Dict[str, Any], store: "MemoryStore") -> Dict[str, Any]:
"""Replay a staged write against the store, bypassing the gate (/memory approve)."""
action = payload.get("action")
target = payload.get("target", "memory")
target_error = _memory_target_error(store, target)
if target_error is not None:
return target_error
if action == "batch":
return store.apply_batch(target, payload.get("operations") or [])
run = _STORE_ACTIONS.get(action)
if run is None:
return {"success": False, "error": f"Unknown staged action '{action}'."}
return run(store, target, payload.get("content") or "", payload.get("old_text") or "")
# =============================================================================
# OpenAI Function-Calling Schema
# =============================================================================
MEMORY_SCHEMA = {
"name": "memory",
"description": (
"Save durable facts to persistent memory that survive across sessions. Memory is "
"injected into every future turn, so keep entries compact and high-signal.\n\n"
"HOW: make ALL your changes in ONE call via an 'operations' array (each item: "
"{action, content?, old_text?}). The batch applies atomically and the char limit is "
"checked only on the FINAL result — so a single call can remove/replace stale entries "
"to free room AND add new ones, even when an add alone would overflow. The response "
"reports current/limit chars and confirms completion; one batch call finishes the "
"update, so don't repeat it. Use the bare action/content/old_text fields only for a "
"single lone change.\n\n"
"WHEN: save proactively when the user states a preference, correction, or personal "
"detail, or you learn a stable fact about their environment, conventions, or workflow. "
"Priority: user preferences & corrections > environment facts > procedures. The best "
"memory stops the user repeating themselves.\n\n"
"IF FULL: an add is rejected with the current entries shown. Reissue as ONE batch that "
"removes or shortens enough stale entries and adds the new one together.\n\n"
"TARGETS: 'user' = who the user is (name, role, preferences, style). 'memory' = your "
"notes (environment, conventions, tool quirks, lessons).\n\n"
"SKIP: trivial/obvious info, easily re-discovered facts, raw data dumps, task progress, "
"completed-work logs, temporary TODO state (use session_search for those). Reusable "
"procedures belong in a skill, not memory."
),
"parameters": {
"type": "object",
"properties": {
"action": {
"type": "string",
"enum": ["add", "replace", "remove"],
"description": "The action to perform (single-op shape). Omit when using 'operations'."
},
"target": {
"type": "string",
"enum": ["memory", "user"],
"description": "Which memory store: 'memory' for personal notes, 'user' for user profile."
},
"content": {
"type": "string",
"description": "The entry content. Required for 'add' and 'replace' (single-op shape). Alias: 'new_text' is also accepted (mirrors old_text)."
},
"old_text": {
"type": "string",
"description": "REQUIRED for 'replace' and 'remove' (single-op shape): a short unique substring identifying the existing entry to modify. Omit only for 'add'."
},
"new_text": {
"type": "string",
"description": "Alias for 'content' (single-op shape). Provided so the replace/remove old_text/new_text pairing works; if both are set, 'content' wins."
},
"operations": {
"type": "array",
"description": (
"Batch shape: a list of operations applied atomically in one call "
"against the final char budget. Preferred when making multiple changes "
"or consolidating to make room. Each item is {action, content?, old_text?}."
),
"items": {
"type": "object",
"properties": {
"action": {"type": "string", "enum": ["add", "replace", "remove"]},
"content": {"type": "string", "description": "Entry content for add/replace. Alias: 'new_text'."},
"new_text": {"type": "string", "description": "Alias for 'content' in a batch op."},
"old_text": {"type": "string", "description": "Substring identifying the entry for replace/remove."},
},
"required": ["action"],
},
},
},
"required": ["target"],
},
}
# Schema text when only one built-in store is enabled: (target description, TARGETS replacement).
_SINGLE_TARGET_TEXT = {
("memory",): (
"The enabled built-in store: 'memory' for personal notes.",
"TARGET: only 'memory' is enabled for personal notes (environment, conventions, "
"tool quirks, lessons).",
),
("user",): (
"The enabled built-in store: 'user' for user profile.",
"TARGET: only 'user' is enabled for user profile facts (name, role, preferences, style).",
),
}
def _build_memory_schema_overrides() -> Dict[str, Any]:
"""Narrow the advertised target surface using the availability snapshot."""
flags = _memory_surface_flags.get()
_memory_surface_flags.set(None)
if flags is None:
flags = get_builtin_memory_store_flags()
targets = [t for t, on in zip(("memory", "user"), flags) if on]
parameters = copy.deepcopy(MEMORY_SCHEMA["parameters"])
target_schema = parameters["properties"]["target"]
target_schema["enum"] = targets
description = MEMORY_SCHEMA["description"]
narrowed = _SINGLE_TARGET_TEXT.get(tuple(targets))
if narrowed:
target_schema["description"], replacement = narrowed
description = description.replace(
"TARGETS: 'user' = who the user is (name, role, preferences, style). 'memory' = your "
"notes (environment, conventions, tool quirks, lessons).",
replacement,
)
return {"description": description, "parameters": parameters}
# --- Registry ---
from tools.registry import registry, tool_error
registry.register(
name="memory",
toolset="memory",
schema=MEMORY_SCHEMA,
handler=lambda args, **kw: memory_tool(
action=args.get("action", ""),
target=args.get("target", "memory"),
content=args.get("content"),
old_text=args.get("old_text"),
new_text=args.get("new_text"),
operations=args.get("operations"),
store=kw.get("store")),
check_fn=check_memory_requirements,
emoji="🧠",
dynamic_schema_overrides=_build_memory_schema_overrides,
)