Files
Agent-Workbench/docs/model-protocols-f-engine-specialists.txt

47 lines
4.3 KiB
Plaintext
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
✔ native text and call IDs preserve original result pairing while unsupported content fails closed (1.1259ms)
✔ one engine child freezes real installation identity before creating tasks and rejects a changed source (13098.4525ms)
✔ owned native tools return saved results, use real approvals/checks and account each actual request once across new engine Runs (40251.2521ms)
✔ forged native scope cannot change a parent file and a final completion claim cannot replace worker proof (20308.5425ms)
✔ duplicate native IDs are rejected without replaying a saved tool (16562.3545ms)
✔ native input exceeding the context budget is retained and sends no model request (15352.2075ms)
✔ stop retains a late reported model usage once and executes no tool from that output (18454.3389ms)
✔ a source change after the original model intent stops before native tools execute (11914.2429ms)
✔ an engine source change while approval is pending invalidates the old permission before any check starts (20706.4131ms)
✔ missing or noncompleted turn end cannot be adopted even with an owned receipt and assistant text (0.6475ms)
✔ closed root with an original live descendant remains unknown and retains its process tree (0.2834ms)
✔ host recovery reserves original pending attempts exactly once and explicit review preserves unknown outcomes (19007.1945ms)
✔ corrupted model reservation cannot reduce recovered budget or permit a new engine request (16877.8397ms)
✔ owned CC Switch SQLite imports the selected TOML Responses wire_api and leaves unknown clients untouched (3234.3774ms)
✔ Responses executes one original tool through Runtime, carries encrypted Items, reports raw usage and seals exact wire bytes (6488.5999ms)
✔ mixed-protocol pool keeps both actual variants and changes route on 429 without re-executing completed tools (4370.0093ms)
▶ Responses incomplete, failed, missing call identity and unsupported hosted outputs never dispatch a tool
✔ Reject {"body":{"id":"resp_owned","status":"incomplete","output":[{"type":"function_call","id":"fc_partial" (3630.0169ms)
✔ Reject {"body":{"id":"resp_owned","status":"failed","output":[{"type":"function_call","id":"fc_failed","cal (2948.9085ms)
✔ Reject {"body":{"id":"resp_owned","status":"completed","output":[{"type":"function_call","id":"fc_bad","cal (3278.7186ms)
✔ Reject {"body":{"id":"resp_owned","status":"completed","output":[{"type":"computer_call","id":"not-supporte (3321.5086ms)
✔ Responses incomplete, failed, missing call identity and unsupported hosted outputs never dispatch a tool (13180.1108ms)
▶ invalid JSON in a nonstream Chat or Responses tool group prevents every tool and pool replay
✔ chat-completions (3471.3136ms)
✔ responses (2630.7028ms)
✔ invalid JSON in a nonstream Chat or Responses tool group prevents every tool and pool replay (6102.4523ms)
✔ wire adapter preserves tool IDs but never passes Responses-private Items to Chat or another endpoint (0.187ms)
✔ connection JSON import accepts explicit Responses protocol and continues to reject unknown protocols without reading client sessions (1186.2766ms)
✔ a stop from the event loop interrupts immediate model replies before later tools (3385.4454ms)
✔ a stop after the first tool preserves a complete interrupted multi-call protocol group (4073.0285ms)
✔ the compatible HTTP loop preserves call IDs, validates errors and waits before executing commands (10411.9691ms)
✔ denial prevents command execution and produces a paired denied result (5347.8428ms)
✔ stopping during approval interrupts pending calls without executing or replaying them (3917.6167ms)
✔ stopping a running fixture command terminates its own process tree and retains uncertain state (3065.8461ms)
✔ command output decodes split UTF-8 and excludes the workbench key from child environment (1458.0358ms)
✔ model-list testing uses a tentative key without changing the saved connection and redacts service errors (6.6736ms)
✔ truncated or repeated model calls fail before execution; tool input validation rejects unknown capabilities (4921.9888ms)
✔ request projection keeps exact tool pairs and complete recent user instructions while preserving disk history (2627.5789ms)
ℹ tests 36
ℹ suites 0
ℹ pass 36
ℹ fail 0
ℹ cancelled 0
ℹ skipped 0
ℹ todo 0
ℹ duration_ms 192697.2863