Four fixes to the tool-search deferral layer, split from PR #92693 (the
availability-cache staleness fix ships separately):
1. The parallel batch planner now peels the tool_call bridge wrapper and
decides admission on the underlying tool — supports_parallel_tool_calls
works again when deferral is active. Unparseable wrappers stay
sequential barriers; bridged calls get exactly the admission the same
call gets direct. tool_search/tool_describe lookups batch concurrently.
2. _short_desc no longer truncates listing lines at 'e.g.', hostnames, or
version strings — a sentence terminator must be followed by whitespace.
3. BM25 indexes the source label (e.g. 'linear' for mcp-linear), so
service-name queries reach tools whose own name omits the service; the
dead 'mcp' prefix token is stripped.
4. Substring-fallback docstring corrected (token misses, not zero-IDF).
Salvaged from #92693 by @alt-glitch with authorship preserved.
* feat(cron): durable failure incidents with signature dedup and ack
Introduce a durable cron incident store (cron_incidents in the shared
cron/executions.db) that groups "same job + same error signature" across
runs, so a known recurring failure stops re-pinging the operator every run
once it has been acknowledged.
- cron/incidents.py: lazily-created incident table (detected -> alerted ->
reviewed -> closed lifecycle; closed is per-signature terminal), sha256
signature dedup over job_id + normalized error, redacted/truncated error
storage, failure-type classification, and ack/list/get/count helpers.
- cron/scheduler.py: record an incident on the failure delivery path and
suppress the per-run failure ping when the exact signature is acked (both
the normal failure path and the processing-raised retry path). Best-effort:
an incident-store error never breaks the cron run or delivery. Streak nudge,
alert-once markers, and delivery-error behavior are untouched.
- hermes_cli: add `hermes cron incidents [--state ...]` and
`hermes cron incidents ack <id>`.
- tests/cron/test_cron_incidents.py: dedup, lifecycle, redaction,
classification, lazy-schema, scheduler gating, and CLI coverage.
Non-goals deferred to later slices: Discord buttons/review view, HMAC action
tokens, owner-agent review launch, approval-gated fixes, incident playbooks.
* refactor(cron): tighten incident lifecycle, wire alerted state and suppressed_acked outcome
Follow-ups on top of the salvaged #94692:
- Drop the dead 'reviewed' state and the SQLite CHECK (state validity
lives in INCIDENT_STATES so future slices can add states without a
table rebuild); lifecycle is detected -> alerted -> closed.
- Actually mark incidents 'alerted' after a failure ping reaches
delivery, on both the normal and exception delivery paths.
- Record ack-suppressed runs with a distinct 'suppressed_acked'
delivery outcome (registered in cron_health monitoring) instead of
the ambiguous generic 'suppressed'.
- Drift-skip alerts explicitly bypass the ack gate (they carry the
remediation command and alert once via drift_alerted already).
- Docs: failure-incidents section in the cron guide.
- Tests for the alerted transition + never-resurrect-closed.
---------
Co-authored-by: Laura López Real <113060513+laulopezreal@users.noreply.github.com>
* refactor(prompt): remove the Nous Subscription block from the system prompt (~1.2K tokens/call)
* chore: retrigger CI (zero-job dispatch failure, auto-heal)
Independent review caught a compaction authority the gate missed:
post-turn micro-compaction (turn_finalizer -> _micro_compact) absorbs the
oldest exchanges into a rolling summary with no pre-compress checkpoint
hook in its path, and both compression.checkpoint_required and
compression.micro_compact could be enabled together — assistant evidence
could vanish into a summary the checkpoint filter later excludes, without
ever reaching the durable provider.
- agent_init: checkpoint_required forces micro-compaction off (warned),
mirroring the native-compaction suppression
- turn_finalizer: defense-in-depth guard at the call site (attribute is
plain mutable state a future path could flip on a live agent)
- behavioral regression test with a sabotage control (gate off proves the
harness reaches the call site; gate armed proves zero calls)
- docs + config example mention the suppression; stale v1 test header fixed
Per review: existing providers should not be retroactively re-versioned or
handed a changed payload. Version 1 is now the implicit historical
on_pre_compress() contract (best-effort, raw message list) that every
pre-existing provider is already on; the fail-closed checkpoint contract
becomes version 2. MemoryManager routes the raw transcript to v1 providers
unchanged and hands the host-normalized evidence list only to v2+ checkpoint
providers, so the plugin surface contract for shipped providers is
byte-identical with the gate off.
The fail-closed gate lived only in compress_context(), but two native
lossy owners compact without ever crossing it (review on #93996):
- codex app-server: in "native"/"off" auto-compaction mode (native is the
default) Hermes preflight is skipped and the codex agent compacts its
own thread inside run_turn() — the compress_context() rejection was
unreachable. init_agent now refuses checkpoint_required together with
api_mode=codex_app_server (BLOCKED_MISSING_PREREQUISITE, extracted as a
testable guard), and run_codex_app_server_turn() fails closed as
defense in depth before a turn can reach the codex-owned boundary.
- Responses server-side native compaction:
native_compaction_context_management() now returns None while the gate
is armed, so context_management never goes on the wire and the
checkpoint-aware Hermes compressor stays authoritative. The suppression
is logged once per process, not silently applied.
Regressions: checkpoint_required + app-server raises before run_turn()
(the session is never created); checkpoint_required keeps
context_management off the wire while the plain configuration still
produces it; the init guard refuses exactly the incompatible pair. Docs
and cli-config.yaml.example describe both bindings.
Refs #93986
Addresses the review on #93996:
- gateway: hygiene and manual /compress load the memory provider only when
compression.checkpoint_required is enabled (skip_memory=not required).
The historical fast path — no provider init, no best-effort hook — is
back for everyone who did not opt in, so default behavior is truly
unchanged.
- conversation_compression: assistant messages carrying both prose and
tool_calls keep their prose in the checkpoint evidence (the tool_calls
payload is stripped, the original message is not mutated); pure
tool-call wrappers without prose are still dropped.
- tests: legacy-database regression proving the _compressed_summary column
is added by the declarative _reconcile_columns() path on a plain reopen
(no version-gated migration needed — append_message works right after),
plus coverage for the prose-preserving filter.
- docs: providers must implement idempotent, content-keyed checkpoint
writes — a fail-closed block means the next attempt re-runs
on_pre_compress over largely the same transcript.
Refs #93986
Context compression is intentionally lossy. Deployments that archive
transcript evidence to an external durable store before compaction had no
way to guarantee the archive actually happened: MemoryManager.on_pre_compress
swallows provider failures by design, so a failed archive silently degraded
into data loss.
This adds an opt-in, provider-agnostic checkpoint contract:
- memory_provider: PRE_COMPRESS_CHECKPOINT_API_VERSION = 1; providers opt in
by advertising pre_compress_checkpoint_api_version. Version 0 keeps the
historical best-effort hook semantics.
- memory_manager: supports_pre_compress_checkpoint() capability probe;
on_pre_compress(require_checkpoint=True) propagates checkpoint-provider
failures and raises when no capable provider completed the checkpoint.
- conversation_compression: new compression.checkpoint_required config key
(default false, documented in cli-config.yaml.example). When enabled,
compaction fails closed with BLOCKED_MISSING_PREREQUISITE (the
uncompressed transcript is preserved) unless a checkpoint-capable provider
confirms the durable checkpoint. Providers receive normalized direct
user/assistant evidence: tool rows, system messages, tool-call wrappers,
and prior compaction summaries are filtered host-side into one stable
contract. codex_app_server compaction is rejected under the gate because
it exposes no truthful pre-compaction transcript boundary.
- hermes_state: persistent _compressed_summary column (declarative schema
migration via _reconcile_columns) so summary provenance survives process
restarts; only the resume model history carries the marker, keeping
get_messages_as_conversation on its existing contract.
- gateway: the lossy hygiene/auto-compact paths load the memory provider
(skip_memory=False) so a required checkpoint also guards those rewrites.
The gate arms only on an explicit boolean True (bare-MagicMock agents in
existing tests have truthy auto-attributes). Default behavior is unchanged:
checkpoint_required=false preserves best-effort semantics for all existing
providers. Contract tests, including a restart round-trip of the summary
marker, in tests/agent/test_pre_compress_checkpoint_contract.py.
Refs #93986
`_background_review_read_before_write_guard` refuses a background-review
`skill_manage` write when the target file was not loaded via `skill_view` in
the same review turn (patch, edit, write_file over an existing file,
remove_file).
`CURATOR_REVIEW_PROMPT` never says so. It lists `skill_view` only under "read
the current landscape", so the reviewer goes straight to the write and every
mutation is refused. The failure is silent from the outside: the curator run
completes, writes nothing, and reads like a pass that simply found nothing to
consolidate. On our deployment that was 32 of 32 attempted writes rejected over
48h before anyone read the logs.
This adds the missing instruction to the toolset block, plus a test that fails
if a future guarded action is added to `skill_manager_tool` without being named
in the prompt — the guard and the prompt have to drift together or not at all.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
The skill_manage guard (added in #55906) refuses any patch/edit of an
existing SKILL.md, or overwrite/removal of an existing support file,
unless the exact target was loaded via skill_view during the review.
Neither _SKILL_REVIEW_PROMPT nor _COMBINED_REVIEW_PROMPT ever mentioned
this, so models routinely issued the write without the pre-read, got
refused, and burned review iterations (#62397).
Both prompts now carry a Read-before-write section scoped to the
guard's actual contract: existing targets only, exact-path pre-read for
support files, transcript quotes don't count, new skills/new support
files exempt, and a bounded one-view-one-retry recovery instead of a
loop. Direction follows #60331 by @kkwills13 with the scope corrections
requested in review (existing-target-only wording, no delete claim,
bounded retry, contract tests for both prompt variants).
Fixes#62397.
Follow-up to the cherry-picked gate-parity fix: the silent 'return 0'
in inject_memory_provider_tools made #81014 undiagnosable — a
configured provider looked half-on with no hint which config key
suppressed its tools. Now an INFO line names the withheld providers
and the gating keys.
The external memory provider's `system_prompt_block()` was injected
unconditionally into the system prompt, while the provider's tools were
gated by `memory_provider_tools_enabled()` via platform_toolsets or
disabled_toolsets. Result: the agent received instructions to call
`mnemosyne_remember`, `mnemosyne_recall`, etc., that did not exist in
its tool surface.
Centralize the gating into `memory_provider_tools_exposed(agent)`, use
it from both `inject_memory_provider_tools` and the system prompt
assembly path, and add regression tests covering:
* memory toolset enabled -> both tools and prompt block exposed,
* memory in disabled_toolsets -> neither exposed,
* memory not in enabled_toolsets and not built-in -> neither exposed,
* the built-in "memory" tool present as an opt-in -> both exposed,
* parity between `inject_memory_provider_tools` and
`memory_provider_tools_exposed`.
Audit finding (Blank Slate): the system prompt advertised web_search,
skill_view, todo, and the hermes-agent skill even when the toolset had
none of them — the model chases phantoms it can't call.
- hermes-agent skill is now essential: cannot be disabled (config reads
strip it, hermes tools writes drop it), cannot be deleted by
skill_manage, is re-seeded past curator suppression, and is seeded
even on .no-bundled-skills profiles (Blank Slate / --no-skills).
- Blank Slate core toolsets grow from file+terminal to
file+terminal+vision+skills: read_file cannot read images and points
at vision_analyze; the essential skill needs skill_view to load.
- HERMES_AGENT_HELP_GUIDANCE degrades to a docs-URL-only variant when
skill tools are absent.
- Execution-discipline guidance drops its web_search lines when web
tools are off (execution_guidance_text renderer).
- Skills-index preamble says 'basic tools like terminal' instead of
naming web_search when web tools are off.
- Coding operating brief drops the todo-tracking sentence when the todo
tool isn't loaded.
All gating keys off agent.valid_tool_names, fixed at session
construction — prompt stays byte-stable per session (cache-safe).
web_extract stopped using an auxiliary LLM long ago (deterministic
truncate-and-store), but browser snapshots still routed oversized
accessibility trees through the auxiliary web_extract model, keeping a
dead-looking aux slot alive across every config/picker surface.
- tools/browser_tool.py: remove _extract_relevant_content and
_get_extraction_model; oversized snapshots always truncate at line
boundaries, store the full tree to cache/web, and append a read_file
pointer (element refs beyond the cut live in the file)
- tools/browser_camofox.py: same — no LLM path
- Remove auxiliary.web_extract slot: config_defaults (removal note, same
pattern as session_search/PR #27590), cli.py defaults + env bridge,
gateway/run.py bridged keys, hermes config display, hermes model picker,
dashboard REST slots, desktop + web AUX_TASKS, i18n labels (en/zh/
zh-hant/ja/ar)
- Docs: env-vars, configuration, fallback-providers, browser + zh-Hans
mirrors (web-search zh-Hans was stale on the old LLM pipeline — synced
to truncate-and-store truth)
- Tests updated: aux bridge uses approval slot, browser tests assert the
LLM path is gone and stored files are secret-redacted
Fixes the four poisoned-connection classes (#81051, #77765, #84132,
#81995) with the SuspectableBackend cheap-mark/lazy-verify contract:
- mark_suspect/ensure_healthy protocol (agent/deadline.py): noticing a
poisoned state never does I/O; the NEXT caller pays once for a health
probe that clears the suspicion or forces a reconnect. A single
teardown-vs-keepalive race or auth-lock corruption can no longer park
a connection permanently — park stays reserved for genuinely
exhausted reconnect budgets.
- keepalive failure marks the connection suspect before requesting
reconnect; the next tool call probes and recycles if unhealthy.
- auth-classified permanent failures on a previously-proven session get
a suspect+reconnect path instead of an immediate park.
- fast-fail (#81995): stdio child pids are tracked at spawn and an
in-flight RPC races a child-watcher task, so a dead subprocess fails
the call immediately with a retryable timeout instead of riding out
the full 300s. Deliberate teardown/reconnect also fails in-flight
calls now instead of leaving them attached to a dying transport.
Dispatch-boundary hardening for test doubles: stubbed sessions
(MagicMock/non-awaitable call_tool, absent child-watcher) fall back to
the exact pre-change inline-await semantics, so only real transports
gain the race guard.
Salvage credit: in-flight approach from #73377 (@luijoc, wedged
transport recovery) and #48069 (@arminanton, keepalive/in-flight
interaction); both PRs' bases predate main's current park/reconnect
architecture, so this is a fresh implementation of their contracts.
Tests: tests/tools/ -k mcp = 639 passed (was 22 new failures during
development; final tree zero).
The pricing snapshot could only express flat per-million rates, so
gemini-3.1-pro sessions with prompts over 200k tokens under-counted
input 2x ($2 vs $4/M) and output 1.5x ($12 vs $18/M).
- Add optional tier fields to PricingEntry: tier_threshold_tokens,
input/output/cache_read_cost_per_million_above (None = flat, falls
back to base rate per-field).
- estimate_usage_cost selects the above-threshold rates for the WHOLE
request once usage.prompt_tokens (input + cache read + cache write)
exceeds the threshold, matching Google's billing semantics.
- Populate gemini-3.1-pro (4.00/18.00/0.40 above 200k; alias
gemini-3.1-pro-preview inherits) and gemini-2.5-pro (2.50/15.00
above 200k).
- Flat entries are untouched: no threshold means no behavior change.
Reported and tier-field shape designed by @tornike14 (#93469).
Tests: below/at threshold unchanged, above-threshold tiered whole-request
pricing, cache-read tier rate and base-rate fallback, preview alias,
flat entries unaffected.
The two local-server context probes in _query_local_context_length read
data.get("max_tokens") as a context-window candidate. On an
OpenAI-compatible /v1/models passthrough max_tokens is the max OUTPUT
tokens, so a 1M-context model advertising a 128K output cap resolves to
128000 and auto-compaction fires ~7x early.
Route both branches through the module's own key vocabulary
(_CONTEXT_LENGTH_KEYS), which already classifies max_tokens as a
_MAX_COMPLETION_KEYS entry.
Local /v1/models probes treated Anthropic `max_tokens` (max output) as the
context window when `max_model_len`/`context_length` were absent. Anthropic
and Anthropic-compatible reverse proxies expose both:
max_input_tokens = context window (e.g. 1M for claude-fable-5)
max_tokens = max output (e.g. 128k)
That under-reported windows (1M → 128k), persisted the wrong value into
context_length_cache.yaml, and fired compression at ~96k (75% of 128k).
Route model objects through a shared helper that prefers input-window keys
via _extract_context_length, and only falls back to max_tokens when no
input-window field is present.
responses.create re-walks the entire request body against the
ResponseCreateParams union graph client-side while holding the GIL.
#93650 documents that walk wedging for 12+ hours on a ~1.4 MB
conversation, starving every other thread including the TTFB/stale
watchdogs — and no socket kill can unblock a pre-network hang.
Hermes payloads are JSON round-trips and already wire format, so the
bulk fields (input, tools) are now routed through extra_body, which the
SDK merges into the JSON body after the transform. Guarded by a
plain-JSON check (anything else keeps the typed path) and a
HERMES_CODEX_SDK_TRANSFORM=1 escape hatch. Applied to both the primary
stream path and the auxiliary adapter.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Follow-ups on top of the salvaged #92785 commit:
- Pre-gate now matches base URLs via normalize_route_base_url and
provider ids via custom_provider_aliases, mirroring the semantics of
get_custom_provider_model_capability. The raw string comparison
silently dropped declarations whose config spelling differed only by
host case or trailing slash (proven empirically: …/v1/ vs …/v1 with a
non-matching provider name returned (False, False) despite an explicit
prompt_caching: true).
- get_provider(..., allow_network=False) in the early-init/stub branch:
the policy runs per request destination (MoA aggregator, auxiliary
replans via blank_cache_policy_stub, early agent init) and a cold
models.dev cache triggered a measured ~450 ms foreground registry
fetch from the send path. A catalog miss degrades to the conservative
side.
- Debug-log the previously silent provider-lookup exception fallback.
- Tests: _make_agent defaults _custom_providers=[] (post-init reality;
keeps built-in-route tests off the catalog/config fallback), the two
early-init tests delete the attr explicitly, and three regression
tests pin the URL-drift, spaced-legacy-name, and no-network contracts
(all three fail on the unfixed commit).
Apply explicit per-model prompt_caching capabilities to custom
chat-completions routes, rather than limiting them to recognized providers,
hosts, or model families.
Keep undeclared routes conservative, derive the marker layout from the wire
transport, and leave Responses and Bedrock caching paths unchanged.
Consolidates the 429-quota-classifier cluster on top of the merged #93419
Anthropic core. Three independent contributor findings salvaged into one
coherent change to the single 429 branch:
- Broaden the 429 usage-limit check from the narrow 'usage limit' string to
the full _USAGE_LIMIT_PATTERNS ('quota', 'limit exceeded', 'key limit
exceeded') and add _BILLING_PATTERNS detection on 429 ('insufficient
credits' wrapped in a 429 instead of 402), guarded by a _RATE_LIMIT_PATTERNS
exclusion so an explicit 'Rate limit exceeded' never promotes to
non-retryable billing. (credit @Pluviobyte, #39441 — earliest submitter)
- Add 'resets in' to the transient signals: Codex's 'Weekly usage limit
reached. Resets in 6hr 29min.' wrongly read as terminal billing because
main only had 'reset in' (no substring match). (credit @LeonSGP43, #63021)
- Add 'reset after' / 'available in' / 'per minute' / 'per second' transient
signals. (credit @jtstothard, #74785)
Supersedes #65633 (defective branch placement, no tests). The aux-client
path already covers these shapes (_is_payment_error catches weekly/quota
walls; _is_rate_limit_error treats 'resets in' as transient), so no change
there.
Tests: 6 new cases (generic quota wall, insufficient-credits 429, rate-limit
guard, Codex resets-in, extra transient phrases). Guard sabotage-verified.
Co-authored-by: Pluviobyte <Pluviobyte@users.noreply.github.com>
Co-authored-by: LeonSGP43 <LeonSGP43@users.noreply.github.com>
Co-authored-by: jtstothard <jtstothard@users.noreply.github.com>
Salvage of #93214 (5 commits squashed onto current main; agent_runtime_helpers.py
diverged since the PR base and was 3-way reapplied). The credential-rotation
guard in recover_with_credential_pool and both restore_primary_runtime paths
only tolerated the custom-naming split when the agent carried the literal label
'custom', so a named custom provider (agent.provider='gemini-no-filter', pool
'custom:gemini-no-filter') tripped the mismatch guard and skipped rotation on
every 401/429. Now all three guard sites use the canonical
credential_pool_matches_provider boundary predicate + resolve_runtime_pool_key,
which recognizes configured named-custom aliases and validates endpoints.
Fixes#93188.
Widened from /review to the class: _build_child_system_prompt now runs
the parent's resolved workspace_path through
agent.prompt_builder.build_context_files_prompt (same discovery/
priority/caps as the main system prompt: .hermes.md > AGENTS.md chain >
CLAUDE.md > .cursorrules; SOUL.md skipped) and embeds the result as
binding conventions. All delegate_task children get it — reviewer
included — since children are built with skip_context_files=True and
previously worked in repos without the repo's own conventions.
The review-engine-local load_workspace_context duplicate is removed;
the reviewer inherits the block via the shared child prompt path.
workspace_path comes only from explicit sources (_resolve_workspace_hint
— TERMINAL_CWD / agent cwd hints, never bare getcwd), so the #64590
install-tree-fallback guard concern doesn't apply.
Tests moved to pin the generalized path (real-filesystem AGENTS.md via
_build_child_system_prompt, empty/no-workspace negatives, reviewer E2E
through start_review). Docs: subagent-context section + /review flow
(en + zh-Hans).
load_workspace_context() resolves the parent's workspace via the same
_resolve_workspace_hint used for child prompts (explicit sources only —
TERMINAL_CWD / agent cwd hints, never a bare getcwd fallback, so the
#64590 install-tree-leak guard concern doesn't apply) and runs it
through agent.prompt_builder.build_context_files_prompt — the exact
discovery/priority/cap logic the main system prompt uses (.hermes.md >
AGENTS.md chain > CLAUDE.md > .cursorrules; SOUL.md skipped). The
result is embedded in the reviewer briefing as binding review
standards. Subagents are built with skip_context_files=True, so without
this the reviewer judged repo work without the repo's own conventions.
5 new tests incl. real-filesystem AGENTS.md discovery through the real
loader. Docs updated (en + zh-Hans).
The reviewer subagent now inherits the primary agent's working skill
context: collect_parent_loaded_skills() gathers launch-preloaded skills
(from the activation notes in ephemeral_system_prompt) and mid-session
skill_view loads (from assistant tool_calls in history), deduped and
capped at 8, and the briefing instructs the reviewer to skill_view each
and treat their conventions as binding for the assessment.
Reference-file reads (file_path=...) don't count as loads; full-skill
injection was rejected as too costly (a single dev skill can be 40KB+).
Docs: delegation.md /review flow updated (en + zh-Hans).
Follow-up to salvaged PR #92767 (review round 2 P1): the .done path
allocated a fresh tail sequence even for items announced earlier via
output_item.added, so a mixed announced/pending stream without
output_index values reordered the calls ([B, A] instead of [A, B]).
First-observed ordering metadata is now recorded for every announced
item and reused at .done; a fresh sequence is allocated only for
genuinely unannounced items. The .done event's own output_index wins
when present, with the announced index as fallback.
Regressions: two announced calls without indices where the first later
receives .done; an announced non-function item preceding a pending call.
Backends that omit per-item done events on a successful completion
(anomalyco/opencode#37159) caused an announced function call to be
silently dropped: the turn ended with output == [] and the tool never
executed. Track calls announced via output_item.added, accumulate
argument deltas, and settle still-pending calls from accumulated state
at a successful terminal event. output_item.done stays authoritative.
Mirrors anomalyco/opencode#43575.
Fixes#93022. A short session (protection window >= transcript) hits the
"insufficient messages" / "no compressible window" branches twice and
permanently trips the anti-thrash breaker, even though nothing was
eligible to compress - compression was never attempted, so there is
nothing "ineffective" to score. The session then rides past the
threshold with no compaction possible (recovery probes only soften,
not fix, the misclassification).
Distinguish "nothing eligible right now" from "attempted and
underperformed":
- New transient _structural_no_op_backoff_until (in-memory, 300s)
armed by _record_structural_no_op() at the three structural no-op
sites: insufficient_messages, no_compressible_window,
empty_post_handoff_window. No strikes accumulate; auto-compaction
resumes on its own once the backoff lapses or the transcript outgrows
the protection window.
- The backoff gates should_compress via
_automatic_compression_blocked_locally and surfaces in
_compression_block_reason as "structural_backoff:<seconds>".
- #40803's frozen-CLI guarantee is preserved: a transcript that can
never shrink retries at most once per backoff window instead of
every turn.
- force=True (/compress) clears an active backoff before attempting;
record_completed_compaction() lifts it - both prove the transcript
is compressible/being worked.
- Genuine attempted-but-underperformed verdicts still strike the
durable ineffective counter unchanged.
Tests: new tests/agent/test_context_compressor_structural_backoff.py;
updated the two tests that asserted the old strike-on-noop behavior.
Address review feedback:
- Add tests for deeply-nested $ref (recursion), top-level JSON array
(already wrapped, no 400 path), and $ref without '#/' prefix (stays
structured).
- Document the deliberate structural (false-positive-tolerant) detection and
its O(n) cost in the helper docstring.
Gemini 3 resolves JSON-Schema $ref/$defs pointers inside a
functionResponse.response payload and rejects unknown references with
HTTP 400 INVALID_ARGUMENT ('referenced name #/$defs/...' does not match
a display_name; see vercel/ai#14369).
tool_describe (and any tool whose result is itself a JSON Schema) returns
schema text that previously went back as a structured response, tripping
Gemini's pointer resolution. Detect such results with a $ref-pointer scan
and wrap them as opaque text instead.
Adds regression tests for the wrap path and the unchanged structured path.
alt-glitch