mirror of
https://github.com/NousResearch/hermes-agent.git
synced 2026-07-31 19:16:29 +00:00
6b81590c55
Systematic prune per AGENTS.md test policy, one pass over every major test tree (gateway, hermes_cli, tools, agent, run_agent, plugins, cli, cron, tui_gateway, honcho/openviking, root-level): - DELETE: source-reading tests (read_text/getsource on prod files), change-detector tests (exact catalog counts, model-name snapshots, config version literals), mock-echo tests (assert a mock returns what it was told), assertion-free/trivial tests, near-duplicate parametrizations (boundaries + one representative kept), async/sync twin duplicates, cosmetic within-file variations. - KEEP (mandatory): security/redaction/approval guards, message-role alternation invariants, prompt-caching/deterministic-call-id invariants, issue-number regression tests (deduped), E2E tests. - 6 test files deleted outright (script-style/no-assert or fully redundant); conftest.py, fakes/, fixtures/ untouched. - tests/acp/conftest.py added: autouse fixture stubs the live models.dev/GitHub/Copilot/Anthropic inventory fetches that ACP server tests performed on every session create — test_server.py 147s → 3.4s, and the tests are now genuinely hermetic. - Sleep-based slowness shrunk where safe (codex_ttfb_watchdog, compression_concurrent_fork, etc.); no wall-clock assertion tightened. Verification: full hermetic suite via scripts/run_tests.sh — 2439 files, 31,130 tests passed, 0 failed, 0 flaky retries, 315s wall (baseline: 583s wall, 13,564s subprocess CPU).
112 lines
3.2 KiB
Python
112 lines
3.2 KiB
Python
"""Tests for AIAgent._summarize_background_review_actions.
|
|
|
|
Regression coverage for issue #14944: the background memory/skill review used
|
|
to re-surface tool results that were already present in the conversation
|
|
history before the review started (e.g. an earlier "Cron job '...' created.").
|
|
"""
|
|
|
|
import json
|
|
|
|
from run_agent import AIAgent
|
|
|
|
|
|
_summarize = AIAgent._summarize_background_review_actions
|
|
|
|
|
|
def _tool_msg(tool_call_id, payload):
|
|
return {
|
|
"role": "tool",
|
|
"tool_call_id": tool_call_id,
|
|
"content": json.dumps(payload),
|
|
}
|
|
|
|
|
|
def test_skips_prior_tool_messages_by_tool_call_id():
|
|
"""Stale 'created' tool result from prior history must not be re-surfaced."""
|
|
prior_payload = {"success": True, "message": "Cron job 'remind-me' created."}
|
|
new_payload = {
|
|
"success": True,
|
|
"message": "Entry added",
|
|
"target": "user",
|
|
}
|
|
|
|
snapshot = [
|
|
{"role": "user", "content": "create a reminder"},
|
|
_tool_msg("call_old", prior_payload),
|
|
{"role": "assistant", "content": "done"},
|
|
]
|
|
review_messages = list(snapshot) + [
|
|
{"role": "user", "content": "<review prompt>"},
|
|
_tool_msg("call_new", new_payload),
|
|
]
|
|
|
|
actions = _summarize(review_messages, snapshot)
|
|
|
|
assert "Cron job 'remind-me' created." not in actions
|
|
assert "User profile updated" in actions
|
|
|
|
|
|
def test_includes_genuinely_new_actions():
|
|
new_payload = {
|
|
"success": True,
|
|
"message": "Memory entry created.",
|
|
}
|
|
review_messages = [_tool_msg("call_new", new_payload)]
|
|
|
|
actions = _summarize(review_messages, prior_snapshot=[])
|
|
|
|
assert actions == ["Memory entry created."]
|
|
|
|
|
|
def test_falls_back_to_content_equality_when_tool_call_id_missing():
|
|
"""If a tool message has no tool_call_id, match prior entries by content."""
|
|
payload = {"success": True, "message": "Cron job 'X' created."}
|
|
raw = json.dumps(payload)
|
|
prior_msg = {"role": "tool", "content": raw} # no tool_call_id
|
|
review_messages = [
|
|
{"role": "tool", "content": raw}, # same content -> stale, skip
|
|
_tool_msg("call_new", {"success": True, "message": "Skill created."}),
|
|
]
|
|
|
|
actions = _summarize(review_messages, [prior_msg])
|
|
|
|
assert "Cron job 'X' created." not in actions
|
|
assert "Skill created." in actions
|
|
|
|
|
|
|
|
|
|
def test_handles_non_json_tool_content_gracefully():
|
|
review_messages = [
|
|
{"role": "tool", "tool_call_id": "x", "content": "not-json"},
|
|
_tool_msg("call_y", {"success": True, "message": "Memory updated."}),
|
|
]
|
|
|
|
actions = _summarize(review_messages, [])
|
|
|
|
assert actions == ["Memory updated."]
|
|
|
|
|
|
def test_empty_inputs():
|
|
assert _summarize([], []) == []
|
|
assert _summarize(None, None) == []
|
|
|
|
|
|
|
|
|
|
def test_removed_or_replaced_relabels_by_target():
|
|
review_messages = [
|
|
_tool_msg(
|
|
"c1",
|
|
{"success": True, "message": "Entry removed.", "target": "user"},
|
|
),
|
|
_tool_msg(
|
|
"c2",
|
|
{"success": True, "message": "Entry replaced.", "target": "memory"},
|
|
),
|
|
]
|
|
|
|
actions = _summarize(review_messages, [])
|
|
|
|
assert "User profile updated" in actions
|
|
assert "Memory updated" in actions
|