fix(gateway): read compaction result flag instead of config flag in hygiene guard - #50098
Closed
srojk34 wants to merge 2 commits into
Closed
fix(gateway): read compaction result flag instead of config flag in hygiene guard#50098srojk34 wants to merge 2 commits into
srojk34 wants to merge 2 commits into
Conversation
…ygiene guard The hygiene auto-compress guard at line 9188 read `compression_in_place` (the config flag — True when the user sets `compression.in_place: true`) instead of `_last_compaction_in_place` (the result flag — True only when archive_and_compact actually succeeded). The hygiene agent is built without a session_db, so in-place compaction always fails silently and `_last_compaction_in_place` stays False. Reading the config flag made the guard think in-place succeeded, triggering `rewrite_transcript()` which replaced the original messages with only the compressed summary — permanent data loss, the exact scenario NousResearch#21301 / NousResearch#49925 was designed to prevent. The normal agent path (line 15887) already reads the correct flag.
alt-glitch
added
type/bug
The /compress command has the same bug: it reads compression_in_place (config) instead of _last_compaction_in_place (result). The tmp_agent is built without session_db, so in-place compaction always fails — reading the config flag bypasses the data-loss guard (NousResearch#44794, NousResearch#39704).
teknium1
added
sweeper:risk-session-state
kshitijk4poor
pushed a commit
to kshitijk4poor/hermes-agent
that referenced
this pull request
Jun 24, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
kshitijk4poor
added a commit
to kshitijk4poor/hermes-agent
that referenced
this pull request
Jun 24, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
kshitijk4poor
added a commit
to kshitijk4poor/hermes-agent
that referenced
this pull request
Jun 24, 2026
teknium1
pushed a commit
that referenced
this pull request
Jun 25, 2026
…guard (#50098) Salvage of #50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
teknium1
pushed a commit
that referenced
this pull request
Jun 25, 2026
In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: #50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
teknium1
pushed a commit
that referenced
this pull request
Jun 25, 2026
pai-scaffolde
pushed a commit
to pai-scaffolde/hermes-agent
that referenced
this pull request
Jun 28, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
pai-scaffolde
pushed a commit
to pai-scaffolde/hermes-agent
that referenced
this pull request
Jun 28, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
pai-scaffolde
pushed a commit
to pai-scaffolde/hermes-agent
that referenced
this pull request
Jun 28, 2026
Contributor
|
Automated hermes-sweeper review: this PR's fix is already present on current Evidence:
Thanks for the focused fix and regression coverage. |
teknium1
added
the
sweeper:implemented-on-main
waefrebeorn
pushed a commit
to waefrebeorn/slermes
that referenced
this pull request
Jul 2, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
waefrebeorn
pushed a commit
to waefrebeorn/slermes
that referenced
this pull request
Jul 2, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
waefrebeorn
pushed a commit
to waefrebeorn/slermes
that referenced
this pull request
Jul 2, 2026
This was referenced Jul 5, 2026
habarmc1223-sudo
pushed a commit
to habarmc1223-sudo/hermes-agent-fluxmem
that referenced
this pull request
Jul 8, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
habarmc1223-sudo
pushed a commit
to habarmc1223-sudo/hermes-agent-fluxmem
that referenced
this pull request
Jul 8, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
habarmc1223-sudo
pushed a commit
to habarmc1223-sudo/hermes-agent-fluxmem
that referenced
this pull request
Jul 8, 2026
santhreal
pushed a commit
to santhreal/hermes-agent
that referenced
this pull request
Jul 13, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
santhreal
pushed a commit
to santhreal/hermes-agent
that referenced
this pull request
Jul 13, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
santhreal
pushed a commit
to santhreal/hermes-agent
that referenced
this pull request
Jul 13, 2026
Gravezzz
pushed a commit
to Gravezzz/hermes-agent
that referenced
this pull request
Jul 21, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
Gravezzz
pushed a commit
to Gravezzz/hermes-agent
that referenced
this pull request
Jul 21, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
Gravezzz
pushed a commit
to Gravezzz/hermes-agent
that referenced
this pull request
Jul 21, 2026
leewenjie
pushed a commit
to leewenjie/hermes-agent
that referenced
this pull request
Aug 7, 2026
…guard (NousResearch#50098) Salvage of NousResearch#50098 by @srojk34, cherry-picked onto current main. The hygiene auto-compress guard and the /compress slash command both read compression_in_place (config flag — is in-place mode enabled?) instead of _last_compaction_in_place (result flag — did in-place compaction actually succeed?). Both agents are built without a session_db, so archive_and_compact always fails silently and _last_compaction_in_place stays False. Reading the config flag makes the guard think in-place succeeded, triggering rewrite_transcript() which replaces the original messages with only the compressed summary — permanent data loss. Co-authored-by: srojk34 <srojk34@users.noreply.github.com>
leewenjie
pushed a commit
to leewenjie/hermes-agent
that referenced
this pull request
Aug 7, 2026
… [2/2] In-place compaction (single durable session id, non-destructive soft-archive) becomes the default. Rotation is now the opt-out fallback via compression.in_place: false. Prerequisite: NousResearch#50098 (hygiene guard reads result flag not config flag) merged first — without it, flipping the default causes permanent transcript loss on gateway hygiene-compress and /compress when no session_db is available. Blast radius (empirically measured on current main): 7 rotation-asserting tests broke and are pinned to in_place=False in the companion test commit: - tests/agent/test_compression_concurrent_fork.py (2) - tests/agent/test_compression_logging_session_context.py (1) - tests/agent/test_compression_rotation_state.py (1) - tests/run_agent/test_compression_boundary_hook.py (2 _make_agent helpers) - tests/gateway/test_compression_concurrent_sessions.py (2) Rotation stays as a working fallback and deserves continued coverage. Plan: .hermes/plans/in-place-compaction-38763.md
leewenjie
pushed a commit
to leewenjie/hermes-agent
that referenced
this pull request
Aug 7, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
/compressslash command both readcompression_in_place(config flag — is in-place mode enabled?) instead of_last_compaction_in_place(result flag — did in-place compaction actually succeed?).session_db, soarchive_and_compactalways fails silently and_last_compaction_in_placestaysFalse. Reading the config flag makes the guard think in-place succeeded, triggeringrewrite_transcript()which replaces the original messages with only the compressed summary — permanent data loss.Changes
gateway/run.py:9188—compression_in_place→_last_compaction_in_placegateway/slash_commands.py:2637—compression_in_place→_last_compaction_in_placeTest plan
test_session_hygiene_preserves_transcript_when_in_place_configured_but_no_db— setscompression_in_place=True+_last_compaction_in_place=Falseon the fake agent and assertsrewrite_transcriptis NOT called