Skip to content

fix(kas): refuse a shell command a denyByDefault spec does not allow - #26501

Merged
KiroCrewCaptain merged 1 commit into
mainfrom
fix/kas-deny-by-default-26437
Oct 31, 2026
Merged

KiroCrewCaptain merged 1 commit into
mainfrom
fix/kas-deny-by-default-26437

Conversation

@bolichen97

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: on KAS, an agent spec's toolsSettings.execute_bash.denyByDefault refuses every shell command its allowedCommands do not match, on every approval mode, so an unattended --approval-mode auto job can no longer run a command the spec never allowed.

Today the key has no KAS rule and is only named in the session-start notice. Every shell call prompts, and an auto-mode cron job approves the prompt: the reporter's mkfifo ran with exit 0. A default-mode job waits about 180 s and is rejected.

Why it matters

A deny-by-default allowlist is how authors confine an unattended agent. Under auto mode it silently became "run anything", and the run reported success.

Not a goal

What changed (motivation → approach → change)

Symptom: an auto-mode job ran an unlisted command. Cause: nothing on the KAS path enforced denyByDefault, and auto mode approves whatever the gate does not deny.

Fix: one new gate row, spec-shell-allowlist, a deny_policy placed after governance and before every grant. It refuses the session agent's own shell call unless the recovered command is plain and one usable allowedCommands entry matches it whole, the same test the spec-builtin-grant row uses. The data travels the path allowedCommands already uses: projection key, session handle, permission event, ToolCall, and the provider event copy in providers/acp_turns.py. It is recorded whatever the governance ceiling says, because it only restricts. A malformed record reads as an empty allowlist, and an entry outside the subset allows nothing. The notice no longer names denyByDefault. The user doc src/kiro_crew/docs/agent-spec-fields.md and the specs in docs/system-specs/modules/ (security.md, acp-client.md, the tier table in memory-skills-hooks.md) describe the row.

Backwards compatibility

Breaking: on KAS, a spec with denyByDefault now has unlisted shell commands refused instead of prompted (or auto-approved). A command listed only by an entry outside the subset is refused too: the reporter's own [a-z]+ entry is one. The notice already names such entries, and ^touch /tmp/probe-allowed-.*\.txt$ is the subset form. Writer sweep on origin/main @ 1bd20d0: git grep -n "denyByDefault" -- src config finds no shipped spec or default that sets it; the only writers are user agent specs.

Reader: src/kiro_crew/llm_helpers_parts/tool_approval.py:_handle_tool_approval -- cron -- AUTO_APPROVE consults permission_floor.refusal_for(security_only=False), which now returns the refusal (test_an_auto_mode_job_refuses_a_command_the_spec_never_allowed)
Reader: src/kiro_crew/hooks.py:HookManager.on_tool_call -- chat -- HOOK_BASED refuses at once (test_an_interactive_session_refuses_it_at_once_instead_of_prompting)
Reader: src/kiro_crew/acp/kas_agents.py:spec_keys_without_carrier -- chat -- the notice drops denyByDefault and still names unusable allowedCommands (test_the_notice_no_longer_names_it, test_an_unusable_pattern_allows_nothing)

Tests

test/test_kas_deny_by_default.py (new): an auto-mode floor refuses the unlisted mkfifo and keeps the listed touch; an interactive gate refuses it at once and still prompts for the listed one; chained, piped, substituted, newline and near-miss commands are refused; an empty list, an unusable entry and an unrecovered command refuse; a spec without denyByDefault, a read, an MCP tool and a sub-session request are untouched; the record is kept under a withholding ceiling, never reaches the wire, follows a mode switch, and a malformed one restricts. test_gate_tiers.py adds the row to the order golden and a scenario; test_hooks.py maps the new field; test_kas_tools_settings.py no longer names denyByDefault as lost.

Proof: not provable -- other: prove.py reverts every production hunk and the new test module then fails at import, which it reports as inconclusive; the per-change reverts below are the proof.

Reverted alone Red in the three test files
gate row 17
handle stamp 15
projection record 16
from_event read 15
record kept under a withholding ceiling 16
plain-command check 1
malformed record reads as () 1

Reproduced on origin/main: the reporter's spec (subset pattern) through permission_floor.refusal_for(security_only=False) approves mkfifo /tmp/probe-notlisted-fifo; with the fix it is refused and touch /tmp/probe-allowed-a.txt is still approved.

Manual verification

N/A -- unit coverage drives the real gate and the auto-mode floor with KAS-shaped permission events built by the real session handle. Windows / macOS unverified (pure-Python change, no platform branch).

Related Issues

Closes #26437

Sibling sweep

Swept: git grep -nE "denyByDefault|fallbackAction" -- src/kiro_crew
Sibling: src/kiro_crew/acp/kas_permissions.py:1531 -- tracked in #26492
Sibling: src/kiro_crew/hook_runtime/gate_tiers.py -- changed here

Pattern harvest

Not generalizable: one spec key with no engine rule, now enforced by one gate row; the other such key is tracked in #26492.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title
  • Existing tests pass and new tests added for new functionality
  • Self-review completed; code follows project style guidelines
  • Documentation updated (if applicable)
  • No secrets, credentials, or internal references in the diff

@bolichen97
bolichen97 requested review from a team and KiroCrewCaptain as code owners October 31, 2026 01:13
@github-actions

github-actions Bot commented Oct 31, 2026 •

Copy link
Copy Markdown
Contributor

Sensitive Change Review

Sensitive files changed:

  • src/kiro_crew/hook_runtime/gate_tiers.py

  • src/kiro_crew/hook_runtime/tool_identity.py

  • src/kiro_crew/hooks.py

  • @KiroCrewCaptain: missing the ## Sensitive change review heading

How to clear this check:

  1. The PR author posts the template below as a PR comment, with
    every field filled (N/A needs a reason). Keep the
    Head commit line: it must name the current head SHA.
  2. The check turns green within a minute, with no push.
  3. A new push resets this check: post it again for the new head.

Only the author's comment counts this way. A reviewer can instead
paste the filled template into an "Approve" review on this head.
Merging still needs a reviewer's approval.

## Sensitive change review
- Head commit: ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a
- What rule changed:
- Before -> after (what was allowed/blocked, now):
- Worst case if wrong (breaks commands / leaks secret / locks user out):
- Evidence checked (which sandbox result proves it):
- No regression:
  - Tools still work (which tools/commands were run, result):
  - Backward compatible (old config / old data / old callers still work, how checked):
  - Existing tests (which suites passed on this head):
- How to undo:

@github-actions github-actions Bot added the readiness: checking Automated validation is still running label Oct 31, 2026
@github-actions

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5) — ⏭️ skipped

Revision ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a touches no user-facing surface (no changes under website/ or committed screenshots), so the UX review was skipped. Does not block merge.

@github-actions

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5) — ✅ PASS

Design-level review of ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

Design-Verdict: PASS

Nothing blocks. The new rule only ever refuses, it covers the auto-approve path the bug came in through, it fails closed on malformed data, and the one behaviour it takes away is listed with its readers.

Notes

  • The row applies only to the agent's own requests. A sub-session or native sub-agent request carries None and is not restricted. That is consistent with each agent having its own spec, but security.md should say so outright so nobody later reads it as a gap.
  • If command recovery from a real KAS run_command event ever comes back empty, this rule refuses every shell call for that spec. That fails closed and a revert undoes it, but one live KAS run of the reporter's spec would confirm that recovery holds beyond the event shapes built in the unit tests.

[DESIGN-REVIEWED] ed15e88

Description read

[DESCRIPTION-READ] bfc185172fc91fafbc2644bda51d334e4e4d9976ec540a8b634d6e1605bb507d

That sha256 names everything this verdict read: the PR description, plus the 1 evidence file(s) the workflow collected from it and handed this review. It covers that evidence because the media strip replaces every attachment URL with the same placeholder, so an edit swapping one attachment for another would leave the description's captured text identical while what this review judged changed. Recompute it with .github/scripts/pr-description-capture.sh over the same inputs; if it differs, the description or that evidence changed after this verdict was formed and any finding drawn from either is unproven.

@github-actions

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5) — ✅ PASS

Premise-level review of ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a — why this exists and whether the shipped surface is the smallest honest version. Updated in place on each push. A BLOCK verdict blocks PR readiness; PASS does not.

First-Principles-Verdict: PASS

Nothing blocks. Each item enforces denyByDefault along the same path allowedCommands already uses (linked issue #26437), and the matcher is a function extracted from the existing one, not a second copy.

What this change ships

Inventory (8 items) — 8 justified

Intent: on KAS, stop an unattended auto-mode job from running a shell command that its agent spec's denyByDefault allowlist never allowed. FIX.

  1. On KAS, a denyByDefault spec refuses any unlisted shell command in every approval mode, auto included — justified
  2. An interactive session now refuses an unlisted command at once instead of prompting — justified
  3. An allowedCommands entry outside the supported subset allows nothing under denyByDefault — justified
  4. A malformed or empty allowlist record refuses everything instead of lifting the restriction — justified
  5. Chained, piped, substituted or unrecovered commands count as unmatched and are refused — justified
  6. The allowlist is recorded even when the governance ceiling withholds grants — justified
  7. A KAS mode switch picks up the allowlist of the agent it moves to — justified
  8. The session-start notice no longer names denyByDefault as unsupported — justified

[FIRST-PRINCIPLES-REVIEWED] ed15e88

Description read

[DESCRIPTION-READ] 3761b11ac4c3132a9478e476903b4aa43f5caca602e7836e4aa1f20e8cd12a79

That sha256 names the PR description this verdict read. Recompute it with .github/scripts/pr-description-capture.sh; if it differs, the description changed after this verdict was formed and any finding drawn from the description is unproven.

@github-actions

Copy link
Copy Markdown
Contributor

GPT 6.1 Review — ✅ no blocking findings

GPT 6.1 completed its review of ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a and found no blocking issues.

This comment is updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] ed15e88

False positive or not applicable? A repository writer can comment:
/ai-review override gpt ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a false-positive: <reason, 20 to 500 characters>

@github-actions

Copy link
Copy Markdown
Contributor

Opus 5.5 Review — ✅ no blocking findings

Reviewed ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a — this comment is updated in place on each push.

Review details

No findings. None of the four candidates held up against the code.

Notes:

  • The refusal text at src/kiro_crew/hook_runtime/gate_tiers.py:1384 always says toolsSettings.execute_bash.denyByDefault, even when the spec set it under the shell key.

[OPUS-REVIEWED] ed15e88

Verdict parsed from the review's SHA-scoped output markers for commit ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a.

False positive or not applicable? A repository writer can comment:
/ai-review override fable ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a false-positive: <reason, 20 to 500 characters>

@github-actions

Copy link
Copy Markdown
Contributor

Security Scope Review -- ✅ PASS

Which legitimate operations does ed15e88f3859ff03137e20e5b8f2a887bd9b0d5a newly refuse? Updated in place on each push. A script-confirmed regression blocks; the model proposes candidates and never decides a refusal.

Verdict basis: adjudicated, zero confirmed regressions.

Security scope review -- adjudicated candidates

  • macos-latest (platform posix): 11 classified, 0 newly refused, 11 still allowed.
  • ubuntu-latest (platform posix): 11 classified, 0 newly refused, 11 still allowed.
  • windows-latest (platform windows): 10 classified, 0 newly refused, 10 still allowed.

No candidate flipped from allowed to refused.

Scope-Verdict: PASS

Reason: The tightening is a new gate row that only fires when a KAS agent's own spec sets toolsSettings.shell.denyByDefault. Every other call carries spec_shell_allowlist=None and is judged exactly as before. The new row is not one of the three shell tiers the classifier runs (SHELL_DENY_TIERS: sensitive-bash, exfil, deny-rules), so the candidates check only that those tiers did not regress. The new row's own over-refusal is not adjudicated and is named below.

TIGHTENING: a new deny_policy row, spec-shell-allowlist, placed after governance and before every grant row (GateTier("spec-shell-allowlist", "deny_policy", _tier_spec_shell_allowlist)) -- it refuses a shell command the spec's denyByDefault does not permit from running unprompted, including in an unattended approval_mode: auto job: "if _command_matches_any(call.command, allowlist): return None / return ToolHookResult.deny_policy("Blocked by this agent's spec: toolsSettings.execute_bash.denyByDefault is set and the command matches none of its allowedCommands")"

ADJUDICATED: 17 candidates, 0 confirmed newly-refused

PLATFORM COVERAGE: I wrote candidates for posix, windows and any.

  • posix: /tmp paths, $HOME, a pipe, and a > redirect. These probe the characters the new row reuses from UNGRANTABLE_COMMAND_CHARS (;&|<>$\(){}`).
  • windows: drive-letter paths with \, a .venv\Scripts\python.exe interpreter, %USERPROFILE%, PowerShell $env:, and a C:\Program Files (x86)\...\git.exe path. That last one carries parentheses, which the plain-command test treats as ungrantable.
  • any: platform-neutral git and make commands, plus kirocrew cron status.

The diff makes no platform-specific path assumption of its own. Its only spelling assumption is the ungrantable character set, which hits Windows spellings hardest ((x86), %VAR% is fine, $env: is not). That case is covered by the windows rows, but only for the base tiers, because the new row cannot be classified.

UNADJUDICATED: The guard is _tier_spec_shell_allowlist in src/kiro_crew/hook_runtime/gate_tiers.py.

Call path: KasHarness projection (kas_agents.projected_shell_allowlist) → AcpSessionHandle.kas_spec_shell_allowlist (re-read on a mode switch through switched_shell_allowlist) → _build_permission_event stamps AcpEvent.spec_shell_allowlist → ToolCall.from_event (_allowlist_patterns) → HookManager.judge GATE_TIERS row → permission_floor.refusal_for under AUTO_APPROVE.

The classifier never sets spec_shell_allowlist, so it cannot reach this row. Narrowings I can argue but cannot confirm:

  1. Plain-command test applied to every listed command. Under a denyByDefault spec, any command containing a character from ; & | < > $ \ ( ) { }is refused even when anallowedCommandsregex would match it whole. kiro-cli's owndenyByDefault` matches the regex against the full command string, so these are newly refused on KAS relative to kiro-cli. Examples:
    • git log --oneline -20 | head -5 with entry git log.*
    • pytest -q > /tmp/out.txt with entry pytest.*
    • echo "$HOME" with entry echo.*
    • "C:\Program Files (x86)\Git\bin\git.exe" status with entry .*git\.exe status
    • powershell -Command Get-Content $env:USERPROFILE\... with any matching entry
  2. Unusable patterns refuse everything they would have covered. An entry outside the literal / escaped-punctuation / .* subset allows nothing. The reporter's own ^touch /tmp/probe-allowed-[a-z]+\.txt$ is an example, as are ls [[:alpha:]]+ and any entry past the 64-entry bound. Under the base code those commands prompted; now they are refused outright. The test file accepts this on purpose (test_an_unusable_pattern_allows_nothing), and the session-start notice names the entry. I'm listing it as the main loss of legitimate operations this row adds.
  3. Leading whitespace and invisible characters. A command with a leading space, a tab, or a non-breaking space is refused ( touch /tmp/probe-allowed-a.txt is in the refused set the tests assert).
  4. Unrecovered command. A shell call whose command Crew cannot recover is refused, even if the engine would have run a listed command.

To probe this row, I would submit each command above with a spec_shell_allowlist holding its natural pattern. The narrowest fix for items 1 and 3 would be to apply the ungrantable-character test only on the grant side (_spec_command_granted). The deny side would then fall back to whole-string regex matching like kiro-cli, while still refusing chained commands that the pattern does not span. That choice belongs to the other lanes; this lane can only name it.

[SCOPE-REVIEWED] ed15e88

@bolichen97
bolichen97 enabled auto-merge (squash) October 31, 2026 01:39
@github-actions github-actions Bot added readiness: passed Eligible automated validation passed for the current revision and removed readiness: checking Automated validation is still running labels Oct 31, 2026

@KiroCrewCaptain KiroCrewCaptain left a comment

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Approved on behalf of the watch holder (bolichen). Design and First Principles pass on this head, UX skipped or pass, no conflict, no open change request.

@KiroCrewCaptain
KiroCrewCaptain merged commit 9f0d02f into main Oct 31, 2026
98 of 99 checks passed
@KiroCrewCaptain
KiroCrewCaptain deleted the fix/kas-deny-by-default-26437 branch October 31, 2026 02:11
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 31, 2026
bolichen97 pushed a commit that referenced this pull request Oct 31, 2026
Backports these main PRs to release/0.11.0, in pick order. Each was a
`git cherry-pick -x` of its squash commit on main:

- fix(memory): read memory.db off the writer lock, split background pool (#24290) (cherry picked from commit fed830c)
- fix(dashboard): bound the pre-prompt build's wait to enter the memory pool (#24414) (cherry picked from commit 520cf2c)
- fix(memory): keep the shared -wal when another process closes the memory store (#25852) (cherry picked from commit dc63bb6)
- fix: read the page-cache prefetch in a child to keep sqlite locks (#26302) (cherry picked from commit 81dde23)
- perf(memory): score facts from a term index and prebuild after restart (#25328) (cherry picked from commit 6824348)
- fix(desktop): retry a failed netstat before Windows recovery gives up (#25792) (cherry picked from commit 043dd9f)
- fix(cli): say a gateway still serves the port when restart's replacement is refused (#25366) (cherry picked from commit 2e13127)
- fix(cli): refuse to stop or restart a gateway the desktop app supervises (#25705) (cherry picked from commit bbf0910)
- fix(hooks): approve kiro-cli read-only file searches after the deny tiers (#25150) (cherry picked from commit a0c0bd4)
- fix(settings): refuse a config save that config.local.json overrides (#26216) (cherry picked from commit 7f8688e)
- fix(workflows): write non-finite floats as null so runs open (#26085) (cherry picked from commit 3591282)
- fix(approvals): give each background tool approval its own id (#25905) (cherry picked from commit 51ef3bb)
- fix(chat): record file changes for kiro-cli v3 write tools (#25650) (cherry picked from commit 5369270)
- fix(sandbox): stop blaming a host-wide OOM kill on a per-tree limit (#25647) (cherry picked from commit 791d2a2)
- fix(acp): replay history when a KAS resume restores no turns (#25521) (cherry picked from commit 47c71cb)
- fix(kas): auto-approve granted built-ins after Crew's own checks (#25339) (cherry picked from commit dba4de5)
- fix(kas): mount web_search as the engine's remote_web_search (#26435) (cherry picked from commit 65c21b5)
- fix(kas): answer a spec's shell.allowedCommands at the gate, after Crew's checks (#25539) (cherry picked from commit 39d07d9)
- fix(kas): relay an MCP server's boolean env flags, never its secrets (#25331) (cherry picked from commit b734a68)
- fix(security): read the -c script past options and redirections (#26087) (cherry picked from commit fea0920)
- fix(kas): refuse a shell command a denyByDefault spec does not allow (#26501) (cherry picked from commit 9f0d02f)
- fix(kas): name a deniedCommands deny aimed at a redirect as not carried (#26489) (cherry picked from commit 4951c0c)
- fix(security): delete the curl-upload deny rule (#26469) (cherry picked from commit 70e206e)

Conflicts resolved by hand:
- #26302: test/spawn_audit_helpers.py adds only the page_cache.py entry
  (the embeddings_child.py entry names a module 0.11.0 lacks);
  test/test_page_cache_prefetch.py takes main's file.
- #25328: test/test_dashboard_server_composition_contract.py lazy-import
  count 93 -> 96 (the three names it adds).
- #25705: test/test_cli_restart_refused_message.py takes main's file.
- #26085: dashboard/handlers/workflows.py adds import math, keeps the
  release typing import.
- #25339: platform/governance.py keeps the release rule for a missing
  ceiling (main's enterprise-baseline clause is not on 0.11.0).
- #26435: harness-parity.md H6 row edited in place on the release text.
- #25539: its acp-client.md and agent-spec-fields.md lines are left out;
  the paragraphs they edit are not on 0.11.0 (security.md carries it).
- #25331: the test and two doc mentions use hoist_managed_servers /
  resume=, the 0.11.0 names of main's renamed hoist_servers_for_session.
- #26087: security package line budget 32,552 -> 32,714 (re-measured).
- #26501: kas_permissions.py comment keeps 0.11.0's shapes (main's
  subagent rules are absent); its acp-client.md line is left out, and the
  agent-spec-fields.md toolsSettings row says denyByDefault is gate-enforced.
- #26489: its agent-spec-fields.md sentence is added to the 0.11.0 row.
- #26469: test_denied_commands_security.py keeps 0.11.0's comment; count 113 -> 112.

Test fixes (host-dependent tests made hermetic; none loosens an assertion):
- tmpdir_helpers.short_tmp_path: a tmp_path override for modules whose code
  judges a path by its components (a .kiro/crew data-home part); an agent
  session's TMPDIR under ~/.kiro/crew made every test path the data home.
  Used by the file-explorer, design-tweak (both), spawn-offload, launch-copy,
  home-pin, native-skill and auto-improvement-members tests; the sage
  ancestor-chain test creates its dir under /tmp for the same reason.
- pi-gate tests and the remote-session Python 3.9 test take the first real
  binary on PATH (playwright_cli_installer_helpers._native_tool_on_path): a
  mise shim under a scratch HOME hangs or writes to stdout.
- snapshot-merge ACL test names its own uid; nobody (65534) is the overflow
  uid inside a user namespace and cannot be named in an ACL.
- installer-step and push_guard interrupt tests spawn with SIGINT at its
  default; a background pytest run starts with SIGINT ignored, which a shell
  trap and Python both keep.

Left out: #23920 (already on release/0.11.0), #26144 (needs
sandbox_launch_mark from #24709), #26088 (fences a main-only port),
#25346 (11 of its tests fail on 0.11.0: the kiro-cli skill scope it
compares against differs there), #26397 (needs #26152's new download
route, not on 0.11.0), #26428 (not merged).
bolichen97 added a commit that referenced this pull request Oct 31, 2026
…26487)

Backports these main PRs to release/0.11.0, in pick order. Each was a
`git cherry-pick -x` of its squash commit on main:

- fix(memory): read memory.db off the writer lock, split background pool (#24290) (cherry picked from commit fed830c)
- fix(dashboard): bound the pre-prompt build's wait to enter the memory pool (#24414) (cherry picked from commit 520cf2c)
- fix(memory): keep the shared -wal when another process closes the memory store (#25852) (cherry picked from commit dc63bb6)
- fix: read the page-cache prefetch in a child to keep sqlite locks (#26302) (cherry picked from commit 81dde23)
- perf(memory): score facts from a term index and prebuild after restart (#25328) (cherry picked from commit 6824348)
- fix(desktop): retry a failed netstat before Windows recovery gives up (#25792) (cherry picked from commit 043dd9f)
- fix(cli): say a gateway still serves the port when restart's replacement is refused (#25366) (cherry picked from commit 2e13127)
- fix(cli): refuse to stop or restart a gateway the desktop app supervises (#25705) (cherry picked from commit bbf0910)
- fix(hooks): approve kiro-cli read-only file searches after the deny tiers (#25150) (cherry picked from commit a0c0bd4)
- fix(settings): refuse a config save that config.local.json overrides (#26216) (cherry picked from commit 7f8688e)
- fix(workflows): write non-finite floats as null so runs open (#26085) (cherry picked from commit 3591282)
- fix(approvals): give each background tool approval its own id (#25905) (cherry picked from commit 51ef3bb)
- fix(chat): record file changes for kiro-cli v3 write tools (#25650) (cherry picked from commit 5369270)
- fix(sandbox): stop blaming a host-wide OOM kill on a per-tree limit (#25647) (cherry picked from commit 791d2a2)
- fix(acp): replay history when a KAS resume restores no turns (#25521) (cherry picked from commit 47c71cb)
- fix(kas): auto-approve granted built-ins after Crew's own checks (#25339) (cherry picked from commit dba4de5)
- fix(kas): mount web_search as the engine's remote_web_search (#26435) (cherry picked from commit 65c21b5)
- fix(kas): answer a spec's shell.allowedCommands at the gate, after Crew's checks (#25539) (cherry picked from commit 39d07d9)
- fix(kas): relay an MCP server's boolean env flags, never its secrets (#25331) (cherry picked from commit b734a68)
- fix(security): read the -c script past options and redirections (#26087) (cherry picked from commit fea0920)
- fix(kas): refuse a shell command a denyByDefault spec does not allow (#26501) (cherry picked from commit 9f0d02f)
- fix(kas): name a deniedCommands deny aimed at a redirect as not carried (#26489) (cherry picked from commit 4951c0c)
- fix(security): delete the curl-upload deny rule (#26469) (cherry picked from commit 70e206e)

Conflicts resolved by hand:
- #26302: test/spawn_audit_helpers.py adds only the page_cache.py entry
  (the embeddings_child.py entry names a module 0.11.0 lacks);
  test/test_page_cache_prefetch.py takes main's file.
- #25328: test/test_dashboard_server_composition_contract.py lazy-import
  count 93 -> 96 (the three names it adds).
- #25705: test/test_cli_restart_refused_message.py takes main's file.
- #26085: dashboard/handlers/workflows.py adds import math, keeps the
  release typing import.
- #25339: platform/governance.py keeps the release rule for a missing
  ceiling (main's enterprise-baseline clause is not on 0.11.0).
- #26435: harness-parity.md H6 row edited in place on the release text.
- #25539: its acp-client.md and agent-spec-fields.md lines are left out;
  the paragraphs they edit are not on 0.11.0 (security.md carries it).
- #25331: the test and two doc mentions use hoist_managed_servers /
  resume=, the 0.11.0 names of main's renamed hoist_servers_for_session.
- #26087: security package line budget 32,552 -> 32,714 (re-measured).
- #26501: kas_permissions.py comment keeps 0.11.0's shapes (main's
  subagent rules are absent); its acp-client.md line is left out, and the
  agent-spec-fields.md toolsSettings row says denyByDefault is gate-enforced.
- #26489: its agent-spec-fields.md sentence is added to the 0.11.0 row.
- #26469: test_denied_commands_security.py keeps 0.11.0's comment; count 113 -> 112.

Test fixes (host-dependent tests made hermetic; none loosens an assertion):
- tmpdir_helpers.short_tmp_path: a tmp_path override for modules whose code
  judges a path by its components (a .kiro/crew data-home part); an agent
  session's TMPDIR under ~/.kiro/crew made every test path the data home.
  Used by the file-explorer, design-tweak (both), spawn-offload, launch-copy,
  home-pin, native-skill and auto-improvement-members tests; the sage
  ancestor-chain test creates its dir under /tmp for the same reason.
- pi-gate tests and the remote-session Python 3.9 test take the first real
  binary on PATH (playwright_cli_installer_helpers._native_tool_on_path): a
  mise shim under a scratch HOME hangs or writes to stdout.
- snapshot-merge ACL test names its own uid; nobody (65534) is the overflow
  uid inside a user namespace and cannot be named in an ACL.
- installer-step and push_guard interrupt tests spawn with SIGINT at its
  default; a background pytest run starts with SIGINT ignored, which a shell
  trap and Python both keep.

Left out: #23920 (already on release/0.11.0), #26144 (needs
sandbox_launch_mark from #24709), #26088 (fences a main-only port),
#25346 (11 of its tests fail on 0.11.0: the kiro-cli skill scope it
compares against differs there), #26397 (needs #26152's new download
route, not on 0.11.0), #26428 (not merged).

Co-authored-by: w <w@k.local>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

2 participants