* feat(agent): coding-context posture with per-model edit-format tuning
Hermes detects when it's running in a coding context — an interactive
surface (CLI, TUI, ACP, desktop) sitting in a code workspace (git repo or
recognised project root) — and shifts into a coding posture. Outside that
(chat platforms, non-workspaces) nothing changes.
The posture is modelled as a frozen RuntimeMode selected from a small
ContextProfile registry (coding/general). A profile is data: the toolset to
collapse to, the operating brief to inject, and seams for model routing and
memory. Every domain reads the same resolved object instead of re-probing
git/config on its own:
- System prompt — RuntimeMode.system_blocks(): an operating brief (gather
context before editing, edit through tools not chat, verify with terminal,
cap retry loops) plus a live git/workspace snapshot, built once and baked
into the stable prompt tier so per-conversation caching is preserved.
- Per-model edit-format tuning — the brief nudges each model family toward
the patch mode it handles best: OpenAI/Codex toward mode='patch' (V4A
multi-file diffs), Anthropic toward mode='replace' (string replacement).
The model id rides on RuntimeMode; unknown families keep neutral wording.
- Skill index — non-coding skill categories are pruned from the prompt's
skill index (discovery-only; skills_list/skill_view still reach the full
catalog, with a disclosure note).
- Toolset — only under the opt-in 'focus' mode does the posture collapse to
the coding toolset + enabled MCP servers; the default posture is
prompt-only and never overrides configured toolsets.
Activation via agent.coding_context: auto (default), focus, on, off.
Subagents inherit the posture for free via toolset inheritance + the shared
prompt builder. Detection is not memoized so a long-lived gateway/TUI
process can't pin a stale posture across working directories.
* feat(agent): cover new-file authoring in the coding edit-format nudge
The per-model edit-format guidance only addressed editing existing code
(patch mode='patch' vs 'replace'), but authoring a brand-new file —
write_file, not patch — is a large fraction of real coding work and the
nudge was silent on it. Surfaced when building a single-file artifact where
the dominant operation was write_file and the steering offered no guidance.
Both family lines now lead with "author new files with write_file; for
edits to existing code prefer ...". Tests assert write_file appears in each
family's brief; unknown families still get neutral wording.
* docs(agent): correct memoization docstring + clarify TUI config-load asymmetry
* feat(agent): sharpen the coding posture — verify-loop facts, wider edit steering, $HOME guard
Tuning pass on the coding posture from dogfooding it as a harness:
- Workspace snapshot now hands the model its verify loop up front:
detected manifests + package manager (lockfile sniff), the exact
verify commands (package.json scripts, Makefile targets,
scripts/run_tests.sh, pytest config), and which context files
(AGENTS.md / CLAUDE.md / .cursorrules) exist at the root. Marker-only
(non-git) projects get the snapshot too instead of nothing. The
"verify before claiming done" brief line was the highest-value piece
in evals — this turns it from advice into an executable loop instead
of making the model rediscover the test command every session. Still
stat-cheap, size-guarded reads, built once at prompt time.
- Edit-format steering covers the families Hermes actually serves:
Gemini and open-weight coding models (DeepSeek, Qwen, Kimi, GLM,
Grok, Hermes, Llama, Mistral, Devstral, MiniMax) steer to
mode='replace' — their RL scaffolds use str_replace-style editors.
Previously only GPT/Codex and Claude families got steering; the
models Hermes users disproportionately run all fell to neutral.
- Operating brief gains four behaviors elite harnesses encode: batch
independent reads/searches in one turn; fix root causes and the bug
class (sibling call paths), not the reported site; no drive-by
refactors/renames/reformatting; never read, print, or commit secrets.
Plus a patch-failure escalation ladder: after the same region fails
twice, rewrite the enclosing function/file with write_file instead of
a third patch attempt.
- $HOME dotfiles guard: a git repo rooted exactly at the home directory
(or a marker sitting in it, e.g. a global ~/AGENTS.md) is user config,
not a code workspace — without the guard, every session anywhere under
a dotfiles-managed home silently flipped to the coding posture. Real
projects under such a home still detect via their own markers/repos;
'on' mode bypasses the guard.
Two additive hardening changes on the interleaved-thinking replay path
introduced by this PR's anthropic_content_blocks channel. Both are scoped
to that channel's blast radius; neither changes correct behavior.
1. Replay-time tool-input re-sourcing (credential safety).
The ordered-block channel captures each tool_use `input` from the RAW
API response in normalize_response, which is NOT credential-redacted.
The parallel tool_calls[].function.arguments IS redacted at storage
time (build_assistant_message, #19798). The verbatim-replay fast path
in _convert_assistant_message replayed the raw block input, so a secret
a model inlined into a tool call (e.g. an Authorization header value
passed inside a terminal command) would ride back onto the wire even
though it is redacted everywhere else in history. Re-source tool_use
input from the redacted tool_calls map by
sanitized id; interleave order (the reason this channel exists) is
unaffected. Adapted from #36071, which re-sources tool inputs the same
way on its replay path.
2. Broaden the thinking-replay 400 classifier (defense-in-depth).
error_classifier only matched "signature" + "thinking", so the
frozen-block variant — "thinking ... blocks in the latest assistant
message cannot be modified. These blocks must remain as they were in
the original response." — carried no "signature" token and fell through
to a non-retryable abort. The anthropic_content_blocks channel prevents
the reorder that triggers this 400 at the source, but if any future
mutator reintroduces it, the turn now self-heals via the existing
strip-reasoning-and-retry recovery instead of crash-looping. A negative
case ensures an unrelated "cannot be modified" 400 (no "thinking") is
not swept in. Mirrors the classifier broadening in #36087 and #36071.
Tests
- tests/agent/test_anthropic_thinking_block_order.py: a replay test
asserting an inlined secret is redacted on the wire while interleave
order is preserved.
- tests/agent/test_error_classifier.py: three cases — frozen-block 400
native and via OpenRouter route to thinking_signature/retryable; an
unrelated "cannot be modified" 400 does not.
Both grafts verified RED (tests fail with the change reverted) then GREEN.
Full adapter, transport, classifier and output-field-leak suites pass.
Co-authored-by: AlexanderBFoley <92330381+AlexanderBFoley@users.noreply.github.com>
HTTP 400 "messages.N.content.M.text.parsed_output: Extra inputs are not
permitted" on the native Anthropic transport. Anthropic SDK 0.87.0 response
blocks carry output-only attributes the Messages *input* schema forbids: text
blocks get `parsed_output` and `citations=None`, tool_use blocks get `caller`.
normalize_response captured blocks verbatim via _to_plain_data and replayed
them as request input on the next turn, so the forbidden fields leaked back ->
400. Like the earlier thinking-block bug, one poisoned turn wedges every
subsequent request in the session (even the diagnostic turn), recoverable only
by switching models or deleting the session.
This is a defect in the anthropic_content_blocks channel added for the
interleaved-thinking fix: it preserved block ORDER correctly but copied every
SDK attribute, including output-only ones.
Fix — whitelist input-permitted fields per block type at all three leak points:
- agent/transports/anthropic.py normalize_response: sanitize at CAPTURE so the
poison never persists to state.db (defence-in-depth).
- agent/anthropic_adapter.py _sanitize_replay_block (new): whitelist used on the
ordered-blocks replay path; also recovers already-poisoned stored sessions.
- agent/anthropic_adapter.py _convert_content_part_to_anthropic: a stored
`text` part is rebuilt from whitelisted fields instead of dict(part) verbatim
(this was the exact content.N.text.parsed_output failure locus).
Whitelist not blacklist, so future SDK output-only fields can't reintroduce it.
Block order and thinking-block signatures are preserved (the reason the channel
exists). Adds tests/agent/test_anthropic_output_field_leak.py; full adapter
suite green (163 tests). Existing poisoned state.db rows scrubbed out-of-band.
Interleaved-thinking turns (adaptive thinking, Claude 4.6+/Opus 4.8) emit
content blocks like:
thinking_1(signed) tool_use_1 thinking_2(signed) tool_use_2
Anthropic signs each thinking block against the turn content preceding it
at its position. normalize_response split the turn into two parallel lists
(reasoning_details + tool_calls), discarding cross-type order, and
_convert_assistant_message rebuilt it as [all thinking][text][all tool_use].
That moved thinking_2 ahead of tool_use_1, invalidating its signature, so
Anthropic rejected the latest assistant message with HTTP 400:
messages.N.content.M: `thinking` or `redacted_thinking` blocks in the
latest assistant message cannot be modified.
Observed repeatedly in agent.conversation_loop against api.anthropic.com /
claude-opus-4-8, recurring across sessions on multi-thinking-block turns.
Fix: carry a verbatim, order-preserving copy of the turn's content blocks
(anthropic_content_blocks) end-to-end - capture in normalize_response,
persist/restore through state.db, and replay unchanged for the latest
assistant message. Gated to turns that actually interleave signed thinking
with tool_use, so normal turns are unaffected.
Adds 3 regression tests including a SQLite round-trip covering the
crash-recovery reload path.
Make Parallel the web search/extract backend with a zero-setup free tier:
- Keyless (no PARALLEL_API_KEY): web_search/web_extract work out of the box via
Parallel's free hosted Search MCP (search.parallel.ai/mcp), and parallel
becomes the default backend when no other web credentials are configured
(ahead of ddgs, which is search-only). A small hand-rolled Streamable-HTTP
JSON-RPC client speaks the MCP's web_search/web_fetch tools; the existing
web_search/web_extract tools are the only tools registered.
- Keyed (PARALLEL_API_KEY set): uses the Parallel v1 REST endpoints
(client.search / client.extract with advanced_settings.full_content) — no beta.
Bumps parallel-web 0.4.2 -> 0.6.0.
- Attribution: on the free path only, results carry provider/attribution and the
CLI tool line reads "Parallel search" / "Parallel fetch"; the paid path is
unbranded.
- Selection/registration: web tools register unconditionally (free MCP backstop)
while check_web_api_key remains a real usability probe; explicit per-capability
backends are honored (so misconfig surfaces) rather than masked by the fallback.
Tested: live web_search/web_extract against search.parallel.ai in keyless and
keyed modes; unit suites for the MCP client, backend selection, and display
labeling; full agent run shows the "Parallel search" label on the free path.
* fix(streaming): stop socket read timeout from preempting stale-stream detector
The stale-stream detector is deliberately scaled to 180-300s so reasoning
models (e.g. Opus) can pause mid-stream during extended thinking. But the
httpx socket read timeout stayed at a flat 120s for cloud providers and fired
first, tearing down healthy reasoning streams before the detector (which owns
retry + diagnostics) could act. Symptom: every Copilot/Opus turn dies with
ReadTimeout at a consistent ~125s and never completes.
Floor the cloud socket read timeout at the stale-stream timeout so it can no
longer fire before the detector. Local providers and explicit
HERMES_STREAM_READ_TIMEOUT / request_timeout_seconds overrides are unchanged.
* test(streaming): pin read-timeout >= stale-stream invariant for cloud reasoning streams
Cover the contract that the httpx socket read timeout is never shorter than
the stale-stream detector for cloud providers on the default: small contexts
floor to 180s, >=50K to 240s, >=100K to 300s; explicit overrides win; local
providers and the unresolved-value fallback are unaffected.
The thinking-signature recovery in agent/conversation_loop.py popped
reasoning_details from messages, then continued to retry. That had two
defects.
First, the strip never reached the wire payload. api_messages is built
once at the start of the turn by shallow-copying every entry in messages
(line 919 area). Each api_messages entry has its own reference to the
same reasoning_details list. When build_api_kwargs runs on every retry
iteration of the inner while-loop, it consumes api_messages, not
messages. Popping reasoning_details from messages left api_messages
untouched, so the retry's request still carried the same thinking
blocks Anthropic had just rejected. The classifier latched
thinking_sig_retry_attempted = True after the first attempt, and the
loop terminated with max_retries_exhausted on the same 400.
Second, the pop mutated the canonical message list. messages is the
same list _persist_session writes to state.db and the session
transcript, so a single recovery permanently wiped every signed
thinking block from the stored conversation. Subsequent turns reloaded
the stripped state, hit the same 400 ('invalid signature' or 'cannot
be modified', see #24107), and the agent stopped responding entirely.
Cascading compaction-ended sessions then chained off the corrupted
parent and the affected chat could not produce a response on any
future turn.
Move the strip onto api_messages, which is the API-call-time list
rebuilt into kwargs on every retry. messages is no longer touched, so
disk I/O stays clean and the recovery actually reaches the wire.
Observed against the native Anthropic Messages API on claude-opus-4-7
and claude-opus-4-8 with the interleaved-thinking-2025-05-14 beta on
hermes-agent 0.12.0 and 0.14.0. PR #24107 narrows the trigger; this
change makes the recovery do what it always claimed to do, and
prevents the destructive aftermath.
Tests cover the api_messages strip in isolation: pop on a shallow copy
does not affect the source, the canonical messages list survives the
strip, idempotency on a duplicate firing path, and a no-op when no
reasoning_details exist on the messages.
Related: #24107, #26959, #17861.
Anthropic returns a 400 when the thinking/redacted_thinking blocks in the
latest assistant message are mutated upstream: 'thinking or redacted_thinking
blocks in the latest assistant message cannot be modified. These blocks must
remain as they were in the original response.'
The classifier's thinking_signature branch only matched on the substring
'signature', so this variant fell through to a non-retryable client error
and hard-aborted the turn -- even though the existing strip-reasoning_details
-and-retry recovery would have healed it.
Broaden the 400 match to also catch 'cannot be modified' / 'must remain as
they were' (still gated on 'thinking'), routing it to the same recovery.
Adds a negative-case test so unrelated 'cannot be modified' 400s are not
swept in.
Defense-in-depth, orthogonal to the root-cause work in #35975 / #17861
(which prevent the block mutation in the first place). Only changes a
terminal-failure into a one-shot recovery.
Signed-off-by: Ian Culling <ian@culling.ca>
Rebased onto current main and re-ported across the restructured
surfaces: model flows now thread confirm_provider/base_url/api_key
through hermes_cli/model_setup_flows.py, the Discord picker lives in
plugins/platforms/discord/adapter.py, and the web dashboard picker
applies chat-mode switches via config.set so the expensive-model
confirmation can ride the response.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
A GPT-5 model rejecting max_tokens returns a 400 whose message contains the
literal substring 'max_tokens' — one of the _CONTEXT_OVERFLOW_PATTERNS. The 400
path in _classify_400 checked overflow patterns before any request-validation
check (which only existed on the 5xx path), so the parameter error was routed
into the compression loop, re-sent with the same bad param, and ended in
'Cannot compress further' on a tiny context.
Hoist a request-validation guard (unsupported/unknown parameter) above the
context-overflow check in _classify_400. Deliberately excludes the generic
invalid_request_error code, which OpenAI also stamps on real overflow 400s, so
genuine overflows still compress. Pairs with the max_completion_tokens param
fix that stops the bad request at the source.
Also adds AUTHOR_MAP entry for the salvaged PR #13902 commit.
Third-party OpenAI-compatible endpoints (self-hosted gateways, OpenRouter,
Azure proxies) fronting gpt-4o / gpt-4.1 / gpt-5+ / o1-o4 models silently
received max_tokens and 400'd with unsupported_parameter, because the three
kwarg-selection sites only checked base_url_hostname(...) == "api.openai.com"
and fell through to max_tokens on every other host. The constraint is
enforced server-side by the model family, not by the URL, so name-based
detection is required as a fallback.
Changes:
- utils.py: new shared helper model_forces_max_completion_tokens(model) that
prefix-matches gpt-4o, gpt-4.1, gpt-5, o1, o3, o4 families on normalized
(lowercased, vendor-prefix-stripped) names.
- run_agent.py: _max_tokens_param ORs the helper into the URL check.
- agent/auxiliary_client.py:
- auxiliary_max_tokens_param gains an optional keyword-only model arg.
- _build_call_kwargs inline branch applies the same check for both
provider == "custom" and non-custom paths.
Tests:
- tests/test_model_forces_max_completion_tokens.py: 31 new cases covering
positive families, negatives (classic gpt-4, claude, llama, mistral, qwen,
deepseek), vendor prefixes, case-insensitivity, whitespace, None/empty,
and substring-not-prefix guards.
- tests/run_agent/test_run_agent.py::TestMaxTokensParam: 5 new model-based
cases (custom + gpt-5.4, openrouter + gpt-4o-mini, custom + o1-preview,
classic gpt-4-turbo keeps max_tokens, llama3 keeps max_tokens).
- tests/agent/test_auxiliary_client.py::TestAuxiliaryMaxTokensParam: new
class, 7 tests covering the URL x model matrix.
* refactor(desktop): dock terminal under chat and simplify file rail
Keep the right rail focused on file browsing while moving the persistent terminal into the chat column bottom slot, and make terminal colors follow the active light/dark mode instead of a fixed Solarized palette.
* fix(desktop): make the terminal a resizable, themed side pane
- Move the terminal into a resizable pane (viewport-% widths) that shares
<main>'s stacking context, so its drag handle no longer sits under the
fixed terminal overlay; works on either rail side.
- Restore +x on node-pty's spawn-helper before the first spawn to fix
"posix_spawnp failed" on macOS prebuilds (real cause; drop the redundant
shell-candidate retry loop).
- Gate terminal open/fit/start on document.fonts.ready and strip leading
blank rows (re-armed before the resize Ctrl-L redraw) so the prompt sits
flush at the top with no starship add_newline gap.
- Inherit the app editor-surface color as the terminal background.
- Bind Ctrl+` (⌃` on macOS) to toggle the terminal; add a palette entry.
* feat(desktop): show platform hotkey hints in the command palette
- Render each palette item's live binding as a <KbdGroup> hint via a new
comboTokens() helper (mac shows ⌘/⌃/⌥/⇧, every other platform shows
Ctrl/Alt/Shift — never a ⌘ on PC).
- Default the terminal toggle to ⌘` / Ctrl+` (the ~ key) on both platforms.
- Drop the hardcoded (⌘⏎) baked into the composer steer tooltip; render it
platform-aware with formatCombo instead.
* fix(desktop): drop the active check on the command-palette terminal item
* fix(desktop): remove active/check states from the command palette
* fix(desktop): allow ⌥/Shift-drag selection over mouse-mode TUIs
Full-screen apps (hermes --tui, vim) enable mouse reporting, so a plain
drag can't select text and ⌘/Ctrl+L (add-selection-to-chat) had nothing
to send. Enable macOptionClickForcesSelection so ⌥-drag on macOS (Shift
elsewhere) forces a native selection over mouse-mode apps.
* feat(desktop): tell the in-pane agent it's embedded in the GUI
Set HERMES_DESKTOP_TERMINAL=1 on the terminal pane's shell env and surface
it in build_environment_hints, so a hermes/--tui launched inside the pane
knows it's next to the GUI chat and that ⌥/Shift-drag + ⌘/Ctrl+L sends a
selection to the composer. Distinct from HERMES_DESKTOP (agent backend).
* refactor(desktop): drop the redundant Ctrl+` terminal-toggle fallback
The toggle now ships as mod+` on both platforms, so the standard combo
index handles it — the bespoke fallback (and its stale 'old default'
comment) is dead weight.
* fix(desktop): read live terminal selection for ⌘/Ctrl+L
A redraw-heavy TUI (spinners/clocks) outruns onSelectionChange, leaving the
React selection state empty so the state-gated shortcut listener never
attached and ⌘L no-op'd. Always listen and read xterm's live selection (with
a native fallback) at press time; only swallow the key when there's text to
send. Drops the now-redundant custom key handler.
* feat(desktop): make any agent aware it's in the Hermes desktop GUI
Generalize the runtime-surface hint: fire for HERMES_DESKTOP (the backend
powering the GUI chat) as well as HERMES_DESKTOP_TERMINAL (a hermes in the
embedded terminal pane), so it's about being inside the desktop GUI, not
about being a TUI. The terminal-pane selection note stays pane-specific.
* feat(desktop): give the GUI agent a read_terminal tool
The in-app terminal buffer lives in the renderer (xterm), so expose it to the
chat agent over the same blocking bridge clarify uses: read_terminal emits
terminal.read.request, the renderer serializes the buffer (visible screen by
default, or a start_line/count range against total_lines) and answers
terminal.read.respond. Gated to the GUI via HERMES_DESKTOP.
Also restores the flipped-layout titlebar inset fix (app-shell +
desktop-controller) for terminal/preview rails at the window's left edge.
* chore(desktop): trim read_terminal comments
* feat(desktop): add a terminal toggle to the statusbar
The file rail lost its terminal icon, leaving ⌘` and the command palette
as the only ways in. Add a one-click toggle to the statusbar's left
cluster, mirroring the command-center item: it reads $terminalTakeover so
it lights up while the pane is open and stays in sync with the hotkey, and
is gated to chat view (the only place the pane can show).
* fix(desktop): relabel the terminal header button to what it does
The in-pane button claimed a focus/split fullscreen toggle ("Focus
terminal view" / "Return to split view", screen-full/normal icons), but
the terminal is just a resizable side pane — there's no fullscreen. The
button only mounts while the pane is open, so the focus branch was dead
and clicking it merely closed the terminal. Relabel to "Hide terminal"
with a close icon, drop the dead conditional and the now-unused takeover
read.
* fix(desktop): move the terminal toggle next to the version item
Relocate it from the left cluster to the right of the statusbar, just
left of the client version item.
* feat(desktop): default the terminal to PowerShell on Windows
Prefer pwsh (7+) then Windows PowerShell 5.1 over cmd.exe, falling back to
comspec only when neither is present. -NoLogo drops the startup banner so
the prompt sits flush like the POSIX shells.
* feat(desktop): show a persistent divider on the terminal pane
The resize sash only painted on hover, so the terminal/chat boundary was
invisible at rest. Add an opt-in `divider` prop to Pane that paints a thin
resting hairline on the resize edge (side-aware, so it tracks the rail when
the layout flips) and enable it on the terminal pane.
* refactor(desktop): resolve the terminal shell instead of hardcoding it
Make shell selection a real resolver: an explicit override wins
(HERMES_DESKTOP_SHELL on both platforms, $SHELL on POSIX), otherwise
auto-detect the best installed shell — pwsh > Windows PowerShell 5.1 > cmd
on Windows, zsh > bash > sh on POSIX. A shared shellSpecFor() picks the
interactive flags by family, so an overridden bash/pwsh/cmd all launch
correctly.
* fix(desktop): repaint the terminal on light/dark switch
Setting term.options.theme updated colors for the DOM renderer but not the
WebGL one, which caches glyph colors in a texture atlas — so already-drawn
cells kept their old palette after a mode switch. Hold the WebglAddon in a
ref and clear its atlas when the theme changes.
* fix(desktop): match the terminal palette to VS Code Light+/Dark+
Adopt VS Code's exact default ANSI palette (the terminalColorRegistry
defaults), enable minimumContrastRatio: 4.5 so foregrounds are clamped
against the background the way the integrated terminal does, and key the
light/dark choice off renderedMode (the painted surface) instead of
resolvedMode so it can't invert. The canvas + inset paint the live skin
surface (--ui-editor-surface-background) so the terminal blends with the
app and follows light/dark, while the contrast clamp keeps colors crisp.
* fix(desktop): tighten command palette search to substring matching
cmdk's default fuzzy scorer matched anything with the query letters
scattered across an item, so e.g. "color" never narrowed to color
entries. Add a substring filter: every typed word must literally appear
in an item's value/keywords, keeping results tight and predictable.
* fix(desktop): blend the terminal header into the skin surface
The persistent-terminal overlay painted the static palette background
(#1e1e1e/#ffffff), so the transparent header strip revealed a near-black
slab above the surface-colored body. Paint the overlay with the live
--ui-editor-surface-background so header and body read as one pane.
* fix(desktop): re-resolve the terminal surface on skin switch
The canvas surface only re-resolved on light/dark change, so switching
skins at the same mode left the WebGL canvas painted with the old tint
until reload. Key the resolve off themeName too. Also trim the palette
comments.
* chore(desktop): drop redundant terminal theming header comment
A binary @file: ref (PDF, docx, spreadsheet, …) expanded to a bare
"binary files are not supported" warning with no content. The model saw a
failure and gave up — e.g. a dropped PDF came back as a text note claiming the
type was unsupported, even though the file was staged on disk right next to it.
Inject an actionable content block instead: the path, mime type, size, and a
nudge to use its tools to read/convert/view the file (and explicitly not to tell
the user the type is unsupported). General across every binary type — not
PDF-specific. The file already resolves where the agent's tools run (local cwd
or the staged copy in a remote session workspace), so it can act on it directly.
New Anthropic models without a recognized version substring (claude-fable-5
and future named/numbered releases) were classified as legacy and routed down
the manual-thinking path, which made OpenRouter emit thinking.type.disabled —
a form reasoning-mandatory Claude models reject with a non-retryable HTTP 400.
Invert the brittle version-substring allowlists to default-to-modern (mirroring
_get_anthropic_max_output): unknown Claude models get the adaptive/xhigh/
no-sampling contract, with an explicit legacy list for older families. Non-Claude
Anthropic-Messages models (minimax, qwen3, …) keep the manual path.
- anthropic_adapter: _supports_adaptive_thinking / _supports_xhigh_effort /
_forbids_sampling_params now default unknown Claude models to modern; legacy
families enumerated in _LEGACY_MANUAL_THINKING_CLAUDE_SUBSTRINGS.
- openrouter profile: omit reasoning entirely (→ adaptive default) instead of
forwarding {enabled:false} for reasoning-mandatory Anthropic models; legacy
Anthropic + all non-Anthropic models still pass the disable form through.
- model_metadata + output-limit table: register claude-fable-5 (1M ctx, 128K out).
Tests assert the invariant ("unknown Claude model -> modern contract; legacy
stays manual; non-Claude unaffected"), not specific model names.
OpenRouter-routed slugs that are absent from models.dev (e.g. a freshly
shipped anthropic/claude-fable-5) fell through to the generic
DEFAULT_CONTEXT_LENGTHS["claude"]=200K entry and under-reported their real
1M window. The step-6 OpenRouter live-metadata fallback was gated on
`not effective_provider`, but an OpenRouter selection sets
effective_provider="openrouter" (inferred from the base URL), so that
branch was dead code for every OR model.
Add a dedicated step-5 OpenRouter branch that consults the live /models
catalog (authoritative, refreshes as new slugs ship) before models.dev and
the hardcoded family defaults — mirroring the existing Nous/Copilot/GMI
branches. Keeps the Kimi-family 32k underreport guard. Per-model values are
respected (claude-haiku-4.5 stays 200K), so it does not blanket-bump to 1M.
Regression tests cover the fable-5 case, the genuinely-200k case, and the
Kimi guard.
LM Studio / llama.cpp-style servers report the context window in tokens
but the prompt size in characters, e.g. "maximum context length is 65536
tokens. However, you requested 65536 output tokens and your prompt
contains 77409 characters". When a provider profile's default_max_tokens
equals the model's context window, the very first request asks for the
whole window as output and the server returns a hard HTTP 400 — even on a
trivial "hi".
parse_available_output_tokens_from_error did not recognise this phrasing,
so the overflow was misrouted to the prompt-too-long/compression path
(which can't help when the input already fits) instead of the output-cap
reduction + retry path. Detect the "requested N output tokens" form,
estimate the input from the character count (~3 chars/token, conservative
so the retried cap stays inside the window), and return the available
output budget so the existing retry logic shrinks max_tokens and succeeds.
The blanket DEVNULL pass muzzled run_oauth_setup_token()'s interactive
'claude setup-token' login, which needs inherited stdin to prompt the
user. Revert that one call and replace the guard's brittle file:line
whitelist with an inline 'noqa: subprocess-stdin' marker that travels
with the code.
When Hermes runs in TUI mode, the gateway child process communicates with
the Node.js parent over a JSON-RPC protocol on stdin. Subprocess calls that
inherit this stdin fd can trigger a race condition where the child's stdin
read returns EOF, causing the gateway to exit cleanly (exit code 0) mid-tool-
execution.
This is the same root cause as issue #14036 (byterover plugin) and PR #39257
(SSH environment backend). This commit applies the fix — stdin=subprocess.DEVNULL
— to all 85 subprocess.run() and subprocess.Popen() calls that execute inside
the TUI gateway child process.
Scope: TUI-context code only (agent/, tools/, plugins/, tui_gateway/server.py).
CLI code (cli.py, hermes_cli/), tests, scripts, and gateway process management
are excluded — they don't run inside the TUI child and inherit the terminal's
stdin, not the JSON-RPC pipe.
85 call sites across 28 files. All files pass syntax check.
hermes auth add openai-codex now creates an independent
manual:device_code pool entry per account instead of routing through
the singleton _save_codex_tokens save path, which collapsed every
added account into the latest login (the second add overwrote the
first account's singleton-mirrored device_code entry). This is the
add-path half of #39236; PR #39243 (already on this branch) fixes the
re-auth half.
manual:device_code entries refresh from their own token pair
(_sync_codex_entry_from_auth_store only adopts the singleton for
source=="device_code"), so they need no providers.openai-codex
shadow. Adding the first credential marks openai-codex active (the
singleton path did this implicitly) so the setup wizard's
get_active_provider() check still passes; subsequent adds leave the
active provider untouched.
Adds SOURCE_MANUAL_DEVICE_CODE constant and a regression test that two
distinct accounts keep distinct token pairs. Updates two existing add
tests to the pool-only behavior.
Co-authored-by: glesperance <info@glesperance.com>
* fix(stream): don't report dropped mid-tool-call streams as output truncation
A streaming tool call whose SSE ends with no finish_reason (the upstream
delivers the tool name + opening '{' then closes the connection cleanly,
no terminator, no [DONE]) was stamped finish_reason='length' by the mock
builder. That routed it through the output-cap truncation path: 3 useless
max_tokens-boosted retries, then the misleading 'Response truncated due to
output length limit' error — even though the model never reported hitting
any cap.
Reproduced live on nvidia/nemotron-3-ultra:free via the Nous dedicated
endpoint, which stalls/drops during large tool-arg generation (50s-4m41s).
Now: when tool args are incomplete AND the provider sent no finish_reason,
tag the response as a partial-stream stub so the loop reports an honest
mid-tool-call drop and asks the model to chunk its output (existing
continuation machinery), instead of escalating output budget and lying.
A provider-reported finish_reason='length' still takes the real-truncation
path unchanged.
* test(stream): update truncated-tool-args test for drop-vs-cap split
test_truncated_tool_call_args_upgrade_finish_reason_to_length pinned the
old behaviour where ANY incomplete tool args → finish_reason='length' with
tool_calls preserved. That single-chunk-no-finish_reason scenario is exactly
the mid-tool-call stream drop now reclassified as a partial-stream stub.
Split into two tests matching the new contract:
- no finish_reason + incomplete args → PARTIAL_STREAM_STUB_ID, tool_calls=None,
_dropped_tool_names set (the drop path)
- explicit finish_reason='length' + incomplete args → tool_calls preserved,
'length' upgrade unchanged (the genuine output-cap path)
quiet_mode was being used to suppress tool-result display when
tool_progress_mode was 'off'. But quiet_mode also gates operational
status messages, so users with /verbose + tool-progress off lost all
status output.
Adds a dedicated tool_progress_mode attribute to AIAgent; the
tool_executor result-rendering path gates on tool_progress_mode != 'off'.
The CLI passes its tool_progress_mode through agent setup and the
tool-progress cycle command syncs it onto the live agent.
Fixes#33860.
Lift the post-loop finalization tail out of run_conversation into
agent/turn_finalizer.py:finalize_turn. Behavior-neutral; run_conversation
4204 -> 3846 LOC, conversation_loop.py 4578 -> 4220.
The region (everything after the main tool-calling while loop): budget-exhaustion
summary, trajectory save, session persist, turn diagnostics, response transforms,
result-dict assembly, steer drain, and the memory/skill review trigger. Lifted
verbatim into a synchronous single-return free function; the 12 post-loop locals
it reads are passed as keyword args and the assembled result dict is returned to
run_conversation (which returns it to the caller). All agent.* side effects fire
exactly as before.
Imports: os + _summarize_user_message_for_log at module top; logger lazy from
agent.conversation_loop (preserves the gateway... err 'agent.conversation_loop'
logger name, no import cycle).
Validation: 1609/1609 tests/run_agent/ pass; live PTY agent turn PASS.
A Responses-API-shaped payload carrying instructions=/input=/store=/
parallel_tool_calls= can reach the native Anthropic messages.stream() /
messages.create() call under a rare api_mode-flip race (e.g. a concurrent
auxiliary vision call mutating a shared agent between the kwargs build and
the stream dispatch). The Anthropic SDK rejects these with a non-retryable
TypeError that kills the whole turn and propagates the entire fallback chain.
Add sanitize_anthropic_kwargs() at both Anthropic dispatch sites: it drops
the Responses-only keys in place and logs a WARNING (with #31673 breadcrumb)
when one is present, so the underlying race stays visible in the wild
instead of being silently papered over.
When agent.interrupt() fires during an active LLM call, the main poll loop
force-closes the worker-local httpx client to stop token generation. That
raises a transport error (RemoteProtocolError) on the worker thread — the
EXPECTED consequence of our own close, not a network bug.
The streaming retry loop misclassified it as a transient connection error
and retried; each doomed retry stalled for the full stream-stale timeout
(up to 300s). Because the gateway caches AIAgent instances per session, the
stale worker outlived the interrupted turn and raced the next turn's request
on shared client state — the root of the multi-minute cascading-interrupt
hang reported in the wild.
Fix: a request-local _request_cancelled token set by the poll loop right
before the force-close, in both interruptible_api_call (non-streaming) and
interruptible_streaming_api_call. The worker's exception handler checks the
token and exits cleanly — no retry, no fallback, no 'reconnecting' status —
instead of treating the forced error as transient. The token is request-
local (not agent._interrupt_requested, which is cleared at turn boundaries)
so a stale worker outliving its turn still recognizes its own forced close.
Original diagnosis and fix by @kristianvast (PR #6600), against the then-
inline methods in run_agent.py. Those were since extracted into
agent/chat_completion_helpers.py, so the fix is reapplied there.
Co-authored-by: Kristian Vastveit <kristianvast@users.noreply.github.com>
A misconfigured/slow external memory provider could hold the agent in
the 'running' state for minutes after the final response was delivered.
MemoryManager.sync_all / queue_prefetch_all looped provider.sync_turn /
queue_prefetch INLINE on the turn-completion path; a provider making a
blocking network/daemon call (a broken Hindsight daemon was observed
blocking ~298s before failing) blocked run_conversation from returning.
Because every interface (CLI, TUI, gateway) marks the agent 'running'
until run_conversation returns, the agent stayed busy for the full block
and any follow-up message triggered an aggressive interrupt that dropped
the message.
Dispatch provider sync/prefetch to a lazily-created single-worker
background executor. sync_all / queue_prefetch_all return immediately;
work completes (or fails, logged) in the background. A single worker
serializes writes so turn N lands before turn N+1. flush_pending()
provides a barrier for session boundaries and deterministic tests.
shutdown_all() drains the executor with a bounded timeout so a wedged
provider can never hang teardown.
Builtin-only / no-provider sessions spawn no executor (zero new threads
in the common case).
A one-off transient transport failure (streaming-close / incomplete
chunked read / 5xx / 408) on an auxiliary LLM call escalated straight to
provider/model fallback (or, for context compression, dropped the summary
and entered cooldown), even when an immediate retry on the same provider
would have succeeded.
Add a single same-target retry at the top of call_llm() and
async_call_llm() — before the existing except-chain — gated on a new
_is_transient_transport_error() that reuses the canonical
_is_connection_error() detector plus a 5xx/408 status check. A second
failure (or any non-transient error: auth, other 4xx, malformed payload)
falls through to first_err and the existing fallback handling unchanged.
This lives in call_llm so every auxiliary task (compression, memory flush,
title generation, session search, vision) shares one transient-retry
surface, rather than each caller re-implementing it. The context
compressor needs no change — it calls call_llm and inherits the retry; its
existing fallback-to-main path (#18458) now composes naturally (retry the
aux model once, then fall back to main only if the retry also fails).
Co-authored-by: ARegalado1 <alberto.regalado@ymail.com>
run_conversation's inner retry loop tracked recovery state in ~15 scattered
bare booleans (per-provider OAuth refresh guards, format-recovery guards,
restart signals). They are now fields on a single TurnRetryState dataclass the
loop mutates in place (_retry.<flag>), giving the recovery bookkeeping a named,
testable home.
Loop-control vars (retry_count, max_retries, max_compression_attempts) stay as
plain locals — they're while-mechanics, not recovery bookkeeping.
Behavior-neutral: pure local→attribute rewrite of 42 references; kwarg NAMES
preserved (e.g. has_retried_429=_retry.has_retried_429). Live simple + tool
turns OK.
Validation: tests/run_agent/ 1615 passed / 0 failed under per-file process
isolation; new test_turn_retry_state.py pins the field contract.
When context compaction rotates agent.session_id, it updates the gateway/tools
session context (set_current_session_id -> HERMES_SESSION_ID env + ContextVar)
but never updates the separate logging session context. The [session_id] tag on
log lines comes from hermes_logging._session_context (set once per turn in
conversation_loop.py), so post-compaction log lines in the same turn carry the
STALE old id while the message/DB/gateway state carry the new one — breaking log
correlation exactly at the compaction boundary.
Call hermes_logging.set_session_context(agent.session_id) alongside the existing
set_current_session_id, guarded so a logging failure can't regress the routing
update. Logs-only; no runtime or caching impact.
Refs #34089
The curator's idle-archival path (apply_automatic_transitions under
prune_builtins) could archive the bundled `plan` skill, killing the
/plan slash command silently — typing /plan then returned 'Unknown
command' with no signal that a skill had vanished. The archived skill's
hash stays in .bundled_manifest, so 'hermes update' wouldn't re-seed it.
Add PROTECTED_BUILTIN_SKILLS ({plan}) enforced at the master gate
is_curation_eligible() (covers archive_skill + the transition walk) and
in the candidate enumerator (so the LLM consolidation pass never sees
them). Immune to prune_builtins, pin state, and LLM judgment.
VolcEngine's api/plan endpoint occasionally leaks raw XML attribute
fragments into tool_use.name when its protocol-translation layer
converts the model's native XML-style tool emission to Anthropic
Messages tool_use blocks, producing names like:
terminal" parameter="command" string="true
execute_code" parameter="code" string="true
session_search" parameter="session_id" string="true
The corruption happens server-side at the provider, but it breaks
every tool call for affected users — no normalization rule in
repair_tool_call can rescue them, so each request runs through three
retries and then aborts as partial.
Add an early sanitizer in agent_runtime_helpers.repair_tool_call that
trims at the first ' " ', " ' ", '<', or '>' character (idx > 0
only) so the rest of the existing repair pipeline (lowercase /
snake_case / fuzzy match) can resolve the cleaned name normally.
Whitespace is deliberately NOT a separator — the legitimate
"write file" -> write_file repair path (covered by
test_space_to_underscore) must keep working.
Tests: 11 new regression cases in TestVolcEngineXmlPollution
covering all three observed polluted names, CamelCase + pollution
mix, single-quote variants, angle-bracket variants, clean-name
passthrough, and the whitespace-preservation guard. All 18 pre-
existing repair tests still pass (29 total in the file).
Replace the ACP-local prefix/suffix matcher + helper with a single
startswith() check against INTERRUPT_WAITING_FOR_MODEL_PREFIX, now
defined once in conversation_loop.py where the sentinel is produced.
Keeps the source of truth in one place so the guard cannot drift if
the status string changes. Net -17 LOC in server.py.
Also add lsaether to release.py AUTHOR_MAP.
The auxiliary Codex adapter maintained its own chat->Responses conversion
loop that forwarded every non-system message's role verbatim into
Responses input[]. When flush_memories()/compression replayed session
history containing assistant tool_calls + role=tool results, those tool
messages leaked into the request and the Responses API rejected them with
HTTP 400: Invalid value: 'tool'.
Route _CodexCompletionsAdapter.create() through the same shared converter
the main agent transport uses (_chat_messages_to_responses_input), so tool
calls become function_call items and tool results become function_call_output
items with a valid call_id. Single conversion path means no future drift.
Also remove the now-dead _convert_content_for_responses() helper — its only
caller was the private conversion loop this change deletes.
Co-authored-by: ProgramCaiCai <techxacm@gmail.com>
Phase 1 of the god-file decomposition plan. run_conversation's ~470-line
once-per-turn setup block (stdio guarding, retry-counter resets, user-message
sanitization, todo/nudge hydration, system-prompt restore-or-build,
crash-resilience persistence, preflight compression, the pre_llm_call hook, and
external-memory prefetch) is moved verbatim into build_turn_context(), which
returns a TurnContext dataclass the loop unpacks.
Behavior-neutral move-and-name refactor: the builder mutates `agent` exactly as
the inline code did; only the locals the loop reads back are returned.
- run_conversation: 4602 -> 4217 LOC (-385)
- agent/conversation_loop.py: 4965 -> ~4580 LOC
- new agent/turn_context.py: focused, dependency-injected, unit-tested in isolation
Tests: tests/run_agent/ 1570 passed / 0 failed under per-file process isolation.
Relocation follow-ups: 413_compression mocks now patch both module references;
nudge/on_turn_start source-inspection guards point at the extracted module.
ContextCompressor inherited a no-op on_session_end() from ContextEngine, so
per-session iterative-summary state (_previous_summary) survived a real session
boundary on a reused compressor instance. Override it to clear the summary the
moment the owning session ends, complementing the point-of-use guard in
compress(). Closes the cross-session contamination path in #38788.
Co-authored-by: dusterbloom <32869278+dusterbloom@users.noreply.github.com>
When a cron or background session compacts, it sets _previous_summary for
iterative updates. If that session ends without /new or /reset (which calls
on_session_reset()), the stale summary survives on the ContextCompressor
instance. A subsequent live messaging session's compaction then injects it as
'PREVIOUS SUMMARY:' into the summarizer prompt — contaminating the live
session with unrelated content from the prior session.
Add an else guard in compress(): when no handoff summary is found in the
current messages but _previous_summary is non-empty, discard it so
_generate_summary() starts fresh instead of iteratively updating a stale
cross-session summary.
Fixes#38788
The per-session compression lock prevents same-window concurrent forks but
not cross-turn ones: the background-review fork shares the parent's
session_id, so if it won a compression race its new child session was never
adopted by the gateway (the fork is single-lifecycle). The next foreground
turn then started from the stale parent and compressed it again, leaving the
same parent with two sibling children.
Set review_agent.compression_enabled = False so the fork never triggers
compression. Both trigger sites in conversation_loop.py gate on
compression_enabled before calling _compress_context, so the fork can never
rotate the shared parent. Review needs full context anyway — compressing
would degrade the memory/skill summary.
The per-session lock is kept as defense-in-depth for any future shared-session
path. Adds a regression test that fails without the flag and passes with it.
Closes#38727
Subagents delegated to a custom endpoint were misrouted when the parent
ran on a different custom endpoint. Both runtimes collapse to
provider="custom", so _resolve_child_credential_pool() treated them as
interchangeable and handed the child the parent's pool. Leasing from it
then overwrote the child's delegated base_url with the parent's endpoint
via _swap_credential() — the child sent the delegated model name to the
wrong endpoint.
Custom runtimes now resolve by endpoint identity (the custom:<name> pool
key derived from base_url). The parent pool is reused only when both
parent and child resolve to the same custom endpoint; unregistered raw
endpoints return None so the child keeps its fixed delegated credential.
Non-custom provider paths are unchanged.
Fixes#7833.
Problem: get_model_context_length() had an early return at the end of the
custom-endpoint probe branch (step 3) that returned DEFAULT_FALLBACK_CONTEXT
(256K) without ever consulting the hardcoded DEFAULT_CONTEXT_LENGTHS catalog
(step 8). Models served through a custom/proxied gateway (e.g. corporate
Anthropic proxy) that didn't expose Ollama or local-server endpoints would
hit this path and get capped at 256K, even when the model name clearly
matched a known entry in the catalog (e.g. claude-opus-4-8 → 1M).
Changes:
- agent/model_metadata.py: Before returning DEFAULT_FALLBACK_CONTEXT at the
end of the custom-endpoint branch, consult DEFAULT_CONTEXT_LENGTHS using
the same longest-key-first fuzzy matching as step 8. Only fall through
to 256K if no catalog entry matches.
- tests/agent/test_model_metadata.py: Updated existing test and added new
test covering the custom-endpoint → catalog fallback behavior.
Fixes#38865
_supports_vision_override() in image_routing.py checked model.supports_vision
and providers.<name>.models, but not the legacy list-style custom_providers
config. A custom provider entry like:
custom_providers:
- name: my-provider
models:
my-model:
supports_vision: true
was ignored, causing image_input_mode=auto to route through the auxiliary
vision_analyze path instead of natively attaching images.
Fix: added a lookup step for custom_providers list entries, matching by
provider name (including 'custom:<name>' variants at runtime).
providers.<name>.models still takes precedence over custom_providers.
13 new tests covering: true/false override, custom: prefix matching,
no-match fallback, non-dict entries, empty lists, models key missing.
When summary_target_ratio is large (e.g. 0.45) and the context_length is
moderate (e.g. 96000), the soft_ceiling (token_budget * 1.5) can exceed
the total transcript size. _find_tail_cut_by_tokens walks the entire
transcript without breaking early, and the resulting compress window is
either empty (compress_start >= compress_end) or a single message whose
summary-of-one overhead saves ~0 tokens.
Both outcomes cause a no-op compression that does not increment
_ineffective_compression_count, so should_compress() returns True on
every subsequent turn and the loop repeats endlessly.
Fix (two layers):
1. _find_tail_cut_by_tokens: when the backward walk consumed the entire
transcript without breaking (cut_idx <= head_end and accumulated <=
soft_ceiling), re-walk with the raw (non-inflated) token budget to
find a meaningful cut that gives the summarizer a useful middle window.
2. compress(): when compress_start >= compress_end, increment
_ineffective_compression_count and log a warning so the existing
anti-thrashing guard in should_compress() can break the loop.
Fixes#40803
* feat(onboarding): opt-in structured profile-build path on first contact
On a user's very first gateway message, Hermes now optionally offers to
build a short profile of them — then, only with consent, gathers durable
facts and persists them to the user-profile memory store (memory tool,
target="user") so future sessions start already knowing who they are.
Inspired by Poke's zero-input onboarding, but consent-first by design:
- The agent OFFERS, never assumes. Declining stops it immediately.
- Before ANY external lookup it states what it will look up and asks.
- It never reads connected accounts (email/calendar) silently — the
exact privacy concern that made naive implementations feel invasive.
Wiring reuses existing infrastructure end-to-end:
- gateway/run.py first-message hook (was a plain self-intro) now swaps in
the profile-build directive when enabled and not yet offered.
- agent/onboarding.py gains profile_build_mode()/profile_build_directive()
+ PROFILE_BUILD_FLAG, latched once via the existing onboarding.seen
mechanism so the offer fires at most once per install.
- config default onboarding.profile_build: "ask" (set "off" to disable).
Added to an existing section, so no _config_version bump needed.
No new storage layer, no new injection path, no prompt-cache impact.
* fix(dashboard): fold onboarding into agent tab to avoid 1-field category
onboarding.profile_build is the only schema-surfaced onboarding field
(onboarding.seen is an internal latch dict), so the dashboard CONFIG_SCHEMA
single-field-category invariant rejected it. Merge onboarding -> agent like
the other small categories.
Compaction summaries now receive the current date and instruct the
summarizer to rewrite completed actions as absolute, dated, past-tense
facts (e.g. "email John about the proposal" -> "Sent the proposal email
to John on 2026-06-07"). A resumed conversation no longer re-issues work
that already happened or treats a finished action as still pending.
The date is resolved via hermes_time.now() (date-only, user-configured
timezone) inside _generate_summary. The compaction summary is a
mid-conversation message that is never part of the cached prefix, so the
date does not affect prompt-cache stability. Date resolution is
best-effort: a clock failure omits the rule rather than blocking
compaction. The rule rides the shared template, so both first-compaction
and iterative-update prompts carry it.
Inspired by Poke's summarization (temporal anchoring + semantic
preservation).
A non-numeric value in env vars like HERMES_STREAM_RETRIES,
HERMES_KANBAN_SPECIFY_MAX_TOKENS, GOOGLE_CHAT_MAX_BYTES, IRC_PORT, etc.
raised ValueError at import/init and crashed startup. Parse them safely,
falling back to the default.
Unified onto the existing utils.env_int(key, default) helper for core/
hermes_cli/tools modules instead of the original PR's three duplicate
local helpers; plugins keep minimal inline guards (no core-utils import).
All existing max()/min()/`or extra.get()` wrappers preserved.
Co-authored-by: annguyenNous <annguyenNous@users.noreply.github.com>
Two isolated reliability fixes:
- chat_completion_helpers: raise on a zero-chunk stream (no finish_reason,
no content/reasoning/tool_calls) so retry handles it instead of
fabricating a successful empty turn.
- model_metadata: parse the OpenRouter/Nous output-cap error phrasing
("maximum context length is N ... (A of text input, B of tool input,
C in the output)") so parse_available_output_tokens_from_error returns
a real cap and the caller stops looping on it.
Salvaged from #40405 (@ashishpatel26) — took the two stream/error-parsing
fixes. The PR also bundled compression-state changes (on_session_start
clearing _previous_summary; cron session-id prefix preservation, #38788);
those touch the compression hot path and are split out for separate review.
Co-authored-by: ashishpatel26 <ashishpatel26@users.noreply.github.com>
