Skip to content

fix(config): report stale session control opt-outs - #10674

Merged
iamwhatever merged 1 commit into
kirodotdev:mainfrom
LuisBrel:fix/conductor-session-control-default
Sep 14, 2026
Merged

iamwhatever merged 1 commit into
kirodotdev:mainfrom
LuisBrel:fix/conductor-session-control-default

Conversation

@LuisBrel

@LuisBrel LuisBrel commented Sep 14, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

A stored agent.session_control: false value can reflect the former default and disable peer-session tools, while Doctor currently gives no clue that the default changed.

Why it matters

Users need a diagnostic path for missing session tools without weakening the supported global security withdrawal.

What changed

  • add agent.session_control to superseded-default diagnostics
  • keep the record report-only so an explicit false value is never rewritten
  • verify detection, read-only loading, Doctor output, and the adoption boundary

No linked issue: this documents a locally reproduced configuration drift path without an existing public issue.

Tests

  • affected superseded-default tests (70 passed, 3 skipped)
  • Black formatting gate
  • isort
  • flake8
  • mypy

Pattern harvest

Rule candidate: regression test
Pattern: a superseded security-sensitive default needs diagnostics without auto-adoption when the former value remains a supported opt-out.

@LuisBrel
LuisBrel requested review from a team and Zedmor September 14, 2026 02:43
@github-actions github-actions Bot added the fork Pull request from a fork (external contributor) label Sep 14, 2026
@LuisBrel

Copy link
Copy Markdown
Contributor Author

Intent: Let Doctor identify a stored session-control value that may reflect a superseded default without changing operator policy.
Not a goal: Automatically enable session-control tools or override an explicit global withdrawal.

@github-actions github-actions Bot added the readiness: action required A blocking check or review needs attention label Sep 14, 2026
@LuisBrel
LuisBrel force-pushed the fix/conductor-session-control-default branch from 1e7094c to 460552e Compare September 14, 2026 02:47
@dwu96

dwu96 commented Sep 14, 2026

Copy link
Copy Markdown
Contributor

👋 Hi! This PR's description is missing some required sections from our PR template. Workflow runs won't be auto-approved until the description is updated.

Missing sections:

  • ## Problem / Motivation
  • ## Why it matters
  • ## What changed
  • ## Tests

Please update your PR description to include these sections, then push or re-save the description. The workflows will be approved on the next cycle.

@github-actions

github-actions Bot commented Sep 14, 2026 •

Copy link
Copy Markdown
Contributor

GPT 5.6 Review (fork) — ✅ no blocking findings

Reviewed 356b818ba1ee702ace99eb6a4b5ccfa4a195c2f0 via the fork AI-review pipeline; updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] 356b818

@github-actions

github-actions Bot commented Sep 14, 2026 •

Copy link
Copy Markdown
Contributor

Opus 4.8 Review (fork) — ✅ no blocking findings

Reviewed 356b818ba1ee702ace99eb6a4b5ccfa4a195c2f0 via the fork AI-review pipeline; updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] 356b818

@github-actions

github-actions Bot commented Sep 14, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Fable 5, fork) — ✅ PASS

Premise-level review of 356b818ba1ee702ace99eb6a4b5ccfa4a195c2f0 via the fork AI-review pipeline — 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/CONCERNS are advisory.

All checks are done. The mechanism is the repo's documented cause-level answer to materialized-default drift, the default flip is corroborated in-tree (CHANGELOG.md:86, same release as the two timeout rows already registered), report-only matches the registry's own rule for a value that doubles as a supported opt-out (sections.py:1299), and I found no unfixed sibling default changes from that release (subagent_max_turns changed its ceiling, not its default). The only unverifiable atom is the literal PR number #8375, which appears nowhere in-tree — like every other row's changed_in.

First-Principles-Verdict: PASS

Verify changed_in="#8375" is the real PR that flipped the default — the number appears nowhere in-tree, and Doctor will print it forever.

What this change ships

Inventory (4 items) — 4 justified

Intent: tell an operator whose pre-flip install materialized agent.session_control: false that their missing peer-session tools may be a stale default, not their choice — a FIX of an unreported drift path (the registry's own docstring calls a missing row "drift goes unreported").

  1. kirocrew doctor now lists a stored agent.session_control: false under Stored Defaults, with the adopt/keep hint — justified
  2. The unsolicited load-path warning now names the same drift until acknowledged — justified
  3. kirocrew config defaults --adopt / --keep can now un-materialize or affirm this key on explicit request — justified
  4. An explicit false is never rewritten automatically (report-only, pinned by the new tests) — justified

Evidence checked: default flip corroborated at CHANGELOG.md:86 beside the #8949/#8891 rows; False documented as the supported global withdrawal at config/sections.py:1288-1300; grep for release siblings found 0 unregistered materialized-default changes; no second mechanism reports drift attribution.

[FIRST-PRINCIPLES-REVIEWED] 356b818

@github-actions

github-actions Bot commented Sep 14, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Fable 5, fork) — ✅ PASS

Design-level review of 356b818ba1ee702ace99eb6a4b5ccfa4a195c2f0 via the fork AI-review pipeline — updated in place on each push. A BLOCK verdict blocks PR readiness; PASS/CONCERNS are advisory.

Design-Verdict: PASS

A correctly-judged report-only registry append: the drift is real (#8375 verified), and an explicit false is the pinned global withdrawal, so not auto-adopting is exactly right.

[DESIGN-REVIEWED] 356b818

Teach Doctor to flag stored false session-control values that may reflect the former default while preserving false as a supported explicit security withdrawal.
@LuisBrel
LuisBrel force-pushed the fix/conductor-session-control-default branch from 460552e to 356b818 Compare September 14, 2026 03:17
@github-actions github-actions Bot added readiness: checking Automated validation is still running readiness: passed Eligible automated validation passed for the current revision and removed readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running labels Sep 14, 2026
@iamwhatever
iamwhatever enabled auto-merge (squash) September 14, 2026 05:08

@iamwhatever iamwhatever left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

Tier 1 auto-approve: fix (2 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, SAST annotations clean, security checklist all-NO, AI reviewers green. Category: registers agent.session_control as a superseded default (False -> True in #8375) so Doctor names a stored False that silently keeps every peer-session tool disabled; the entry is deliberately report-only and pinned out of the adopting set, with a test asserting the stored value and the on-disk config are left untouched, so an operator's explicit withdrawal is never overridden. CodeQL is not applicable on this fork PR (default-setup emits no check-run); SAST coverage is Semgrep only, latest run success with 0 annotations.

@iamwhatever
iamwhatever merged commit 7b51129 into kirodotdev:main Sep 14, 2026
84 checks passed
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Sep 14, 2026
amadsalmon pushed a commit to amadsalmon/KiroCrew that referenced this pull request Sep 22, 2026
Teach Doctor to flag stored false session-control values that may reflect the former default while preserving false as a supported explicit security withdrawal.

Co-authored-by: LuisBrel <193017400+LuisBrel@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

fork Pull request from a fork (external contributor)

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants