Commit Graph
17 Commits
Author SHA1 Message Date
Devika Verma eeb3c04d91 fix(gui): the turn summary counts changes that worked, and pairs results by call id (OPE-238)
Review of the first version against 366 recorded runs found four defects and one
semantic gap, all fixed here.

- Two same-named calls in one batch swapped results, exit codes, origins and audit
  links in the live view: the finished event was matched to the LAST pending item by
  name. `tool_proposed` now carries the tool call id and the GUI matches on it.
- A command's exit status was read from the 300-character result preview; a long
  command pushed it out, and a timed-out command showed a green dot. The exit code,
  time-out and background flags now ride on `tool_finished` and on the persisted
  display sidecar.
- After a reload every non-denied call replayed as "ok", so a failed write showed as
  done. The status is persisted with the result and replayed.
- A notice inside a turn (compaction, continuation, a saved memory) split the summary
  in two and showed the first half while the turn was still running. One block per
  turn now, emitted after the answer.

The header reads "N changes auto-approved" and the count means "things that changed
state and that nobody approved": the engine stamps each cardless call's effect
("change" or "read", from its risk class and tier) and the summary counts failed
attempts and read-only actions apart. Several edits of one file collapse into one row;
background commands are labelled; after 25 rows the list offers the rest.

`turn_summary` gains `failed` and `reads`, and per-action effect, tier and shell
outcome; docs/headless.md describes the fields.
2026-10-29 14:08:30 +05:30
Devika Verma c3c813349b feat(gui): show the changes auto-approve made without asking, at the end of a turn (OPE-238)
A user who was away had no simple view of what ran with no approval card. After a
turn in which auto-approve let actions through (a gate tier or a reviewer "allow"),
the transcript now ends the turn with a collapsed "N changes made without asking"
block: file changes with diffs, commands with their exit status, other actions,
each with who let it through and a link to its audit rows.

- engine: one TURN_SUMMARY event just before the event that ends the turn, listing
  tool call id, tool, origin, status and the reviewer's reason (no arguments).
  Wraps run / retry / resume; closing the wrapper closes the turn as before.
- audit: /v1/audit takes call_id; Activity opens filtered to one call, clearable.
- gui: tool items carry their call id, live and on reload; the block is built from
  the persisted approval origins, so a reloaded session shows the same summary.
- docs/headless.md: the event in events.jsonl.
2026-10-29 14:08:30 +05:30
Devika Verma f44f760dbc feat(permissions): secret-looking data leaving the machine always needs a person (OPE-236)
Whether a model declines to upload a .env file is model behaviour, not policy,
and OpenWorker lets users pick any model. A new floor, placed with the other
human-only floors (ahead of every allowlist, the allowed-sites wall and the
reviewer): an action that sends data off the machine (egress or external risk:
web tools, connector writes, MCP tools; or a shell command running a network
client) AND names a secret-looking file (.env*, *.pem, id_rsa*, credentials*,
*token*, ...) or carries a secret-looking value (private-key headers, sk-,
AKIA, ghp_, xox?- shapes, api_key = <long value>) is human-only in every
asking mode and in bypass-approvals; dangerous-bypass clears it and records
the floor. Templates and public keys (.env.example, *.sample, *.pub) are not
secrets. A backstop on top of the sandbox network allowlist: shell egress
can't be recognised completely.

Corpora: gate-104 (browser upload of an SSH key) moves from reviewer_eligible
to human_only (secure stays hard_deny); four secret-egress gate rows are
appended (gate-133..136). Seven frozen legacy reviewer rows the floor now
intercepts (danger-003/010, inject-001/002/008/014/016) stay as they are; the
hygiene test lists them explicitly and checks they're caught by this floor.
Docs: the floor joins the safety-check list; the Windows sandbox page notes
what "Allow everything" leaves to the approval flow.
2026-10-29 11:24:12 +05:30
Devika Verma a952d5463b run: --no-web-tools keeps web_search and web_fetch out of the tool list (OPE-226)
For sealed evaluation runs and air-gapped work. A network block only refuses the call: the model still sees both tools, can spend turns on them, and the record cannot say they were absent. The flag withholds them at registration (build_engine(web_tools=False)), and summary.json and the trajectory record web_tools: false. Without the flag nothing changes.

Tests: with the flag both calls get the engine's unknown-tool answer and the record says false; without it web_fetch is present, shown with a non-URL the tool rejects before any network I/O.
2026-10-20 17:35:20 +05:30
Rohit C Prasad 9de6b83c7f run: retry a failed model call after 2 s, then back off
Waits are now 2, 5, 10, 30, 60, 120, 240, 300, 300 s; the first was 30 s.
One dropped connection no longer costs half a minute.
2026-10-05 11:20:53 -07:00
Rohit C Prasad 5c957d2185 run: the retry line gives the cause, and a stop signal ends the wait between retries
A failed model call prints what was underneath it, such as a proxy's 403.
SIGTERM during the wait before a retry now stops the run and leaves its record (exit 130).
2026-10-04 15:46:50 -07:00
Rohit C Prasad bab5b07c9b run: connectors are available, and the record uses the new names
A run has the connectors connected on this computer, as the app does.
summary.json and the trajectory say coworker, approval_mode and auto_answer.
2026-10-04 11:41:00 -07:00
Rohit C Prasad 6e0d58199a run: a folder request is asked in the terminal and yes gives the folder
Yes, yes but read only, no, or a different folder. With --auto-answer or no terminal it is declined by rule, as before.
2026-10-04 11:20:39 -07:00
Rohit C Prasad f3bae90322 run: --approval-mode replaces --mode, and the earlier flag names are gone
--mode, --persona, --attendance and the unlisted mode names are no longer accepted.
Flags cannot be shortened, so --mode is refused instead of being read as --model.
2026-10-04 11:18:40 -07:00
Rohit C Prasad 900bc9f7f2 run: Ctrl-C at a prompt ends the run and leaves its record
Exit code 130. The prompt is read on a daemon thread so the run does not wait for Enter to exit.
2026-10-03 21:05:49 -07:00
Rohit C Prasad 25c9e0ba2d run: uses this computer's settings, asks in the terminal, and saves the session
Default mode is ask; approvals and questions are separate (--auto-answer); --isolated keeps the sealed benchmark run.
Without --out the record goes to sessions/<workspace path>/<session id> in the state folder; --allow-site adds sites for the run.
Exit code 0 only when the task finished; 3 when it stopped early with its record written.
2026-10-03 21:05:07 -07:00
Rohit C Prasad 14e5e8195f run: the approval mode that asks is named ask
The stored name, interactive, is still accepted but unlisted.
2026-10-03 20:47:17 -07:00
Rohit C Prasad 07cc4a3caa run: the task can be piped in on standard input
With no --prompt or --prompt-file, a piped standard input is the task; a terminal or an empty pipe stops with exit code 2.
2026-10-03 20:37:49 -07:00
Rohit C Prasad 8064af1ed1 run: --coworker chooses the coworker, and the workspace defaults to the current folder
--persona still works, unlisted. An unknown coworker id stops the run with exit code 2.
2026-10-03 20:04:17 -07:00
Devika Verma 31fac7049a Rename openworker run --root to --add-dir (OPE-196)
"root" already means the superuser and the top of the filesystem; `--add-dir` says what
the flag does and is the name the best-known agent CLIs use for the same idea. Recorded
in summary.json as args.extra_dirs.
2026-10-02 15:11:22 +05:30
Devika Verma 1d0073faaf Add --root to openworker run: extra writable folders beside the workspace (OPE-196)
A harness can put its output contract outside the workspace: CyberGym-E2E runs the agent
in /src and grades /output/fix.patch and /output/poc.bin. OpenWorker's file tools only
write under a declared root, in every permission mode (the dangerous mode clears the
permission floor, but the tools still resolve paths against the session's roots by
construction), while the shell is not scoped. So without a way to declare /output, a
delivery depended on which tool the model happened to pick.

`--root DIR` (repeatable) adds writable roots after the workspace, which stays the folder
relative paths resolve against; the folders are announced to the model in the session's
directories block and recorded under args.roots in summary.json. The reference harness
does the same for Gemini CLI with --include-directories /src,/scripts,/output.
2026-10-02 15:11:22 +05:30
Devika Verma 78acad6758 Add openworker run: one task, no human, a record left behind (OPE-196)
A headless one-shot command beside `up`: give it a task, a folder and a model; it builds the same engine the desktop app uses, runs to the end with attendance `auto` (coworker/unattended.py), and leaves a record: every event, the final conversation, a summary with tokens and cost, every answer the engine gave on the absent user's behalf, the audit log, a per-call model log, and a trajectory in the ATIF interchange format. The trajectory and the exit code are the stable contract; the other files are internal.

State and scratch folders are pointed under --out before the engine starts, so a run never touches the machine's own conversations, keys or settings. A transient provider failure or an empty reply is waited out with backoff and a nudge instead of ending the run; permanent errors are not retried. Tool results are bounded with the full text spilled beside the record. A small price table gives an estimated cost per call as a cross-check; an unpriced model gets counts but no cost.

The mode axis decides approvals (`dangerously-bypass-approvals` prints its warning line here); the runner adds no policy of its own.
2026-10-02 15:11:21 +05:30