Skip to content

fix(members): no briefing upkeep for crewmates without file tools - #21430

Merged
KiroCrewCaptain merged 1 commit into
kirodotdev:mainfrom
rubencu:fix/briefing-item-no-file-tools
Oct 20, 2026
Merged

KiroCrewCaptain merged 1 commit into
kirodotdev:mainfrom
rubencu:fix/briefing-item-no-file-tools

Conversation

@rubencu

@rubencu rubencu commented Oct 16, 2026 •

Copy link
Copy Markdown
Contributor

Problem / Motivation

Goal: A crewmate whose template mounts no tool that writes files is told it cannot keep its briefing, instead of being told to keep it with file tools it does not have.

Item 6 of the crewmate protocol ([HOW YOU WORK], _MEMBER_BRIEFING_ITEM in src/kiro_crew/context_assembly/member.py) tells every crewmate to keep its [CURRENT ASSIGNMENT] briefing "with your file tools". build_member_section picks that wording with a platform check only (member_briefing_supported); it never asks what the crewmate's spec mounts. Every conductor template mounts no file-writing tool on purpose (agent_materialization/conductor_agents.py: "no tool that can write a file, not fs_write and not code"), and mounts execute_bash for named scripts, never auto-approved.

Seen on a 0.10.0.1 install with three crewmates bound to kirocrew-conductor:

  • sponsored-services-captain and career-coach keep members/<slug>/briefing.md through execute_bash (python3, cp): about 80 shell calls between them from 10-13 to 10-16. They ran unprompted only because that host's safety override auto-approves. On a default install each write is an approval prompt.
  • kirocrew-maintainer has never written one: it has no members/kirocrew-maintainer/ directory.

docs/request-for-change/rfc-crewmate-dashboard-upkeep-skill.md §1 already records that "a crewmate with no file tools cannot write it".

Why it matters

Every conductor-template crewmate is given, on every injected turn, an instruction it can follow only by writing a file through the shell, which its template exists to prevent, or not at all. On a default install that shell write waits on an approval prompt, which an unattended patrol cycle cannot answer.

Not a goal

  • Giving these crewmates a way to write the briefing (a mate tool, a dashboard editor, a file tool on the template). No kirocrew-mate tool writes it: update_self takes a name, a description and a face.
  • Changing what is read. Layer 4 keeps reading a briefing already on disk for every crewmate, on every platform that supports it.
  • Changing the Windows wording (_MEMBER_BRIEFING_ITEM_UNAVAILABLE), which still applies wherever layer 4 is never injected.
  • Editing Crewmate self-configuration via a kirocrew-member MCP; retire triggers, session_color, route_crew, select_crew #19240's design table, which lists briefing.md as "already self-writable".

What changed (motivation → approach → change)

A crewmate is told to keep its briefing only when its own spec can write files. The member section asks that spec, and words item 6 and the briefing layer from the answer:

The crewmate's spec Item 6 [CURRENT ASSIGNMENT] header
mounts a file writer (the default spec, the worker, most templates) unchanged: keep it "with your file tools" unchanged: "yours to maintain, injected from <path>"
mounts none (every conductor template) the template gives it NO tool that writes files, so it does not maintain the file or write it by any other means, and keeps its working memory in the DM thread; the briefing may be out of date, the thread wins where the two disagree, and a stale one is reported to the user "not yours to maintain, injected from <path>; only the user edits or deletes it, so it may be out of date"; the empty placeholder carries no invitation to write one
on a platform without layer 4 (Windows) unchanged _MEMBER_BRIEFING_ITEM_UNAVAILABLE unchanged

A no-write crewmate's briefing is therefore deliberately the user's. One written before this change, as sponsored-services-captain and career-coach did through the shell, stays injected as it is, marked as possibly out of date. The user updates or clears it by editing or deleting <data home>/members/<slug>/briefing.md (the default data home is ~/.kiro/crew); the crewmate's Notes tab shows it read-only.

The answer comes from spec_has_file_write(spec) and binding_has_file_write(binding), new in src/kiro_crew/member_mate_tools.py beside binding_has_mate_tools. They read the same spec through the same _read_template. A spec writes files when it mounts fs_write (by that name, its write alias or its KAS ids str_replace and fs_append), code, * or @builtin, once excludedTools is subtracted. That is the reading spec_grants_tool_search uses, with the shared mount, exclusion and alias readings in platform.tool_names. The writer names follow governance's filesystem.write class (BUILTIN_TOOL_SCOPES). A row that binds no template is read as kirocrew, the spec it runs. A spec that cannot be read keeps today's wording. The shell does not count, and the new text names no tool to write the file with. docs/architecture/context-management.md describes the new variant.

Backwards compatibility

Compatible: only prompt wording changes, and only for a crewmate whose spec mounts no file-writing tool. Every other crewmate gets byte-identical item 6 and layer 4 text. No API, stored data or file format changes.

Removes nothing: every crewmate still has its briefing read and injected. A crewmate without file tools loses only the instruction to write the file, so one that wrote it through the shell stops doing so and its briefing becomes the user's to edit. No code reads the layer's header (on main, CURRENT ASSIGNMENT appears only in this builder and the marker scrubber's patterns in context_assembly/markers.py).

Tests

New test/test_member_briefing_file_tools.py:

  • test_spec_has_file_write_reads_tools_then_exclusions: 22 spec shapes, including * as the whole value, @builtin, KAS ids (str_replace and fs_append count, delete_file does not), alias and family exclusions, and shell or read-only tool lists that do not count.
  • test_the_writer_names_follow_governance_and_the_alias_table: the writer set equals governance's filesystem.write class, and every short spelling in platform.tool_names.KIRO_TOOL_ALIASES counts exactly when the tool it names writes (write does; shell and read do not).
  • test_binding_has_file_write_reads_the_spec_the_member_runs: a reader spec answers False and an editor spec True; a missing spec, a traversal name, a non-string name and an unreadable file keep today's wording.
  • test_a_row_that_binds_no_template_is_read_as_the_default_one: a row with no kiro_agent is read as kirocrew, the template member_template_id runs for it.
  • test_a_conductor_crewmate_is_told_it_cannot_write_its_briefing: runs the real kirocrew-conductor installer over a default spec that mounts fs_write and code, binds a crewmate to the result and renders its section: the new item 6, the briefing still read under a header that names the file, says only the user edits or deletes it and that it may be out of date, no "with your file tools", item 7 intact.
  • test_a_briefing_the_member_cannot_write_is_marked_as_possibly_stale: on the conductor template an existing briefing is still injected, and item 6 says it may be out of date, that the thread wins, and that a stale one goes to the user.
  • test_an_empty_briefing_is_not_an_invitation_to_write_one: the empty placeholder on the conductor template.
  • test_a_template_that_mounts_fs_write_keeps_the_upkeep_instruction: a template with fs_write keeps _MEMBER_HOW_YOU_WORK and the yours to maintain, injected from <path> header byte for byte.
  • test_the_platform_gap_outranks_the_tool_gap: with layer 4 unsupported, the conductor crewmate gets the Windows wording, not the new one.
  • test_the_no_write_item_sends_working_memory_only_to_the_thread: the new item names the DM thread and no shell or mate tool.

prove.py --base origin/main on this head: PROVEN (with the production hunks reverted, an assertion fails). Determinism: the new file is green 20 of 20 runs at -n0, 10 of 10 at -n 4 --dist loadgroup --max-worker-restart=2 beside test_member_mate_tools.py and test_member_prompt.py, and in a shuffled order (seed 145).

Local gates on this head: black, isort, flake8, mypy, comment-history, agent-SDK boundary, brand, harness parity, feature map, file-line cap, docs-lint, sync-IO, loop-bound locks, testpaths coverage, memory-store seam, the flake-pattern ratchet, semgrep on the diff, git diff --check. Related backend set from scripts/run_scoped_tests.py --surface backend on this head: 32,269 passed, 18 failed and 22 errors, and the same 40 node ids fail on a clean origin/main worktree at 999fcd1 on this host. All 40 are in test_acp_spawn_offload.py, test_dashboard_gallery_dispatch_tree.py, test_dashboard_gallery_real_tree.py and test_auto_improvement_members.py: this host's pytest temp root sits under the Crew data home, which the essential-source checks refuse as a project resource. The frontend related set was not run: the diff touches no file under website/.

Manual verification

  • On this install, read-only, spec_has_file_write over the installed specs: False for kirocrew-conductor, kirocrew-ledger-conductor, kirocrew-pipeline-conductor and kirocrew-security-conductor; True for kirocrew and kirocrew-worker. config.json binds career-coach, kirocrew-maintainer and sponsored-services-captain to kirocrew-conductor.

Not verified: a live crewmate turn on a running gateway showing the new text; the rendering test builds the same section from the real installed conductor spec.

Related Issues

Related: #19240, whose field table lists briefing.md as "already self-writable"; for a crewmate on a conductor template it is not. #21404 (open) also edits src/kiro_crew/member_mate_tools.py; this branch is based on main and adds its two functions apart from that PR's hunks, so either can land first.

Pattern harvest

Rule candidate: review-prompt
Pattern: prompt text that tells an agent to use a capability ("with your file tools") is chosen without asking the agent's spec whether it mounts that capability; the review should ask which templates receive the text and whether each one can follow it.

Checklist

  • At most two commits (one is the norm), with a Conventional Commits title (feat|fix|docs|style|refactor|perf|test|chore|ci|build|revert: ...)
  • 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

@rubencu
rubencu requested review from a team and KiroCrewCaptain as code owners October 16, 2026 23:55
@rubencu
rubencu requested a review from cruisercohen October 16, 2026 23:55
@github-actions github-actions Bot added fork Pull request from a fork (external contributor) readiness: action required A blocking check or review needs attention readiness: checking Automated validation is still running and removed readiness: action required A blocking check or review needs attention labels Oct 16, 2026
@github-actions

github-actions Bot commented Oct 17, 2026 •

Copy link
Copy Markdown
Contributor

UX Review (Opus 5.5, fork) — ⏭️ skipped

Revision bef4d30cdab044898145f194156cbe41ed3c489b touches no user-facing surface (no changes under website/ or committed screenshots), so the UX review was skipped. Advisory — does not block merge.

@github-actions

github-actions Bot commented Oct 17, 2026 •

Copy link
Copy Markdown
Contributor

First Principles Review (Opus 5.5, fork) — ✅ PASS

Premise-level review of bef4d30cdab044898145f194156cbe41ed3c489b 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.

First-Principles-Verdict: PASS

Before merge, check one thing: on claude and codex, which ignore a spec's tools, conductor crewmates still have file writers but are now told to drop briefing upkeep.

What this change ships

Inventory (7 items) — 7 justified

Intent: stop telling conductor-template crewmates to keep a briefing file they have no tool to write. This is a FIX.

  1. A conductor crewmate's protocol item 6 now says it cannot write its briefing and should keep working memory in the DM thread — justified
  2. The [CURRENT ASSIGNMENT] header for such a crewmate now reads "not yours to maintain… only the user edits or deletes it" — justified
  3. The empty-briefing placeholder no longer invites such a crewmate to write one — justified
  4. Item 6 tells such a crewmate the thread wins over a stale briefing and to report a stale one to the user — justified
  5. A spec that cannot be read keeps today's upkeep wording — justified
  6. New spec_has_file_write / binding_has_file_write in member_mate_tools.py, each with one consumer — justified
  7. context-management.md now describes the no-write variant of layers 2 and 4 — justified

[FIRST-PRINCIPLES-REVIEWED] bef4d30

Description read

[DESCRIPTION-READ] 0c5cb9bba97549f0daf0dad796490aed9fcff16983e1829652328f997def19b2

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

github-actions Bot commented Oct 17, 2026 •

Copy link
Copy Markdown
Contributor

Design Review (Opus 5.5, fork) — ✅ PASS

Design-level review of bef4d30cdab044898145f194156cbe41ed3c489b 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

Check that conductor crewmates lose nothing they need when a fresh session starts: with no briefing upkeep, the DM thread is their only working memory.

[DESIGN-REVIEWED] bef4d30

Description read

[DESCRIPTION-READ] 30541f31ccc5d2d662c0fb38d6351d7afd3a68cc404ecdf180811504080b1e80

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

github-actions Bot commented Oct 17, 2026 •

Copy link
Copy Markdown
Contributor

GPT 6.1 Review (fork) — ✅ no blocking findings

Reviewed bef4d30cdab044898145f194156cbe41ed3c489b via the fork AI-review pipeline; updated in place on each push.

Review details

No findings.
[GPT-REVIEWED] bef4d30

@github-actions

github-actions Bot commented Oct 17, 2026 •

Copy link
Copy Markdown
Contributor

Opus 5.5 Review (fork) — ✅ no blocking findings

Reviewed bef4d30cdab044898145f194156cbe41ed3c489b via the fork AI-review pipeline; updated in place on each push.

Review details

No findings.

[OPUS-REVIEWED] bef4d30

@github-actions github-actions Bot added readiness: action required A blocking check or review needs attention and removed readiness: checking Automated validation is still running labels Oct 17, 2026
@rubencu
rubencu force-pushed the fix/briefing-item-no-file-tools branch from ee72f2e to a8c833c Compare October 19, 2026 18:51
@rubencu

rubencu commented Oct 19, 2026

Copy link
Copy Markdown
Contributor Author
  • Briefings that already exist become permanent span=5786d62c1361

fixed at a8c833c, by the second branch of "Clears when": the PR now states that a no-write crewmate's briefing is deliberately the user's, and names how it is cleared.

  • The layer's header names the file and says whose it is: [CURRENT ASSIGNMENT — not yours to maintain, injected from <path>; only the user edits or deletes it, so it may be out of date].
  • Item 6 tells the crewmate to treat that section as possibly out of date, that the DM thread wins where the two disagree, and to tell the user when it is stale.
  • The user updates or clears it by editing or deleting <data home>/members/<slug>/briefing.md (default data home ~/.kiro/crew); the crewmate's Notes tab shows it read-only. The description ("What changed", "Backwards compatibility") and docs/architecture/context-management.md (the layer table, the briefing paragraph and the capability row) say so.
  • Not taken: skipping injection for a no-write spec. The PR's "Not a goal" keeps layer 4 read for every crewmate on every platform that supports it.
  • test_a_briefing_the_member_cannot_write_is_marked_as_possibly_stale and test_a_conductor_crewmate_is_told_it_cannot_write_its_briefing pin the header and item 6.

@rubencu

rubencu commented Oct 19, 2026

Copy link
Copy Markdown
Contributor Author
  • The "do not write it by any other means" line stops crewmates that keep their briefing through the shell today

intended. Every conductor template mounts no tool that can write a file on purpose ("not fs_write and not code either", agent_materialization/conductor_agents.py), and mounts execute_bash for the bundled acceptance evaluator, never auto-approved. A shell write of briefing.md routes around that charter, and on a default install each one is an approval prompt an unattended patrol cycle cannot answer.

  • What those crewmates keep: a briefing already on disk stays injected, under a header that names the file and says it may be out of date; their working memory moves to the DM thread; the user edits or deletes <data home>/members/<slug>/briefing.md.
  • Giving them a way to write it (a mate tool, a file tool on the template) is listed under "Not a goal".

@rubencu

rubencu commented Oct 19, 2026

Copy link
Copy Markdown
Contributor Author

Rebased onto main at 02eeb00 and pushed a8c833c with the Design fix (dispositions above).

The two reds on ee72f2e were not from this diff: Backend Tests (3.12, 7) timed out in test/test_session_allocation_bindings.py::test_every_name_patched_through_the_facade_reaches_every_owner_reading_it (#21671, fixed on main by #21640), and Coverage Gate failed closed on that shard. The new base carries #21640 and #22178.

@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author
  • "Do not write it by any other means" for bash-only templates that are not conductors and auto-approve shell writes

needs-a-decision, for a maintainer. What the PR does and the facts behind it:

  • No shipped template is in that set. The only shipped specs that mount execute_bash without fs_write or code are the conductor specs in agent_materialization/conductor_agents.py, and none of them ships an auto-approval for it. The case is an operator-authored template.
  • That operator keeps today's wording by mounting fs_write (or code) on the template: binding_has_file_write reads the template's own tools and excludedTools, and item 6 and the [CURRENT ASSIGNMENT] header are then byte-identical to main.
  • The open question: should a template that auto-approves execute_bash in its own allowedTools, and mounts no file writer, count as able to keep its briefing? The PR says no: it counts governance's filesystem.write class (fs_write, code) and not the shell. Saying yes is a small change to spec_has_file_write (count execute_bash and its shell alias when the spec's own allowedTools grants it). The conductor specs as shipped keep the no-write wording either way, because none ships that grant.

@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author
  • Two crewmates that keep their briefing through the shell will stop updating it, so it freezes as the user's note

needs-a-decision: a maintainer's review of this PR is the confirmation the lane asks for. The trade is stated in the description ("What changed", "Backwards compatibility") and in docs/architecture/context-management.md.

  • What those crewmates keep: the briefing stays injected, under a header that names the file and says only the user edits or deletes it and that it may be out of date. Item 6 makes the DM thread win where the two disagree and has them tell the user when it is stale.
  • What they lose: shell writes of briefing.md. Their template mounts no file writer on purpose, and on a default install each such write is an approval prompt an unattended patrol cycle cannot answer.
  • If a maintainer would rather they keep a briefing, the route is a write path for crewmates without file tools (a mate tool, or a file tool on the template), which this PR lists under "Not a goal". docs/request-for-change/rfc-crewmate-dashboard-upkeep-skill.md §1 records the same gap.

@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author

PR Hygiene on a8c833c did not run its checks: its 10-minute job timeout cancelled it while actions/checkout was still fetching (run 39306017378, job 118357489024: "The job has exceeded the maximum execution time of 10m0s"). The other three jobs in that run make the same full fetch and passed (Automated Rule Check, SAST, Inclusive Language).

Could a maintainer re-run the failed job (gh run rerun 39306017378 --failed)? A fork contributor can't, and a push or a body edit would queue new runs that need approval again.

@github-actions github-actions Bot added the merge conflict Branch has merge conflicts with its base — author must resolve before merge label Oct 20, 2026
Item 6 of the crewmate protocol told every crewmate to keep its briefing
"with your file tools", chosen only by a platform check. A crewmate bound to a
conductor template mounts no tool that writes files, by design, so it could
follow that only by writing the file through the shell or not at all.

member_mate_tools gains spec_has_file_write / binding_has_file_write, read from
the spec the crewmate runs: true when fs_write (or its write alias), code, "*"
or @Builtin is mounted after excludedTools. When it is false, the member section
renders a third item 6 that names the gap and sends working memory to the DM
thread. Layer 4 keeps reading the briefing, which is now the user's: its header
names the file and says it may be out of date, and item 6 says the thread wins
where the two disagree and to tell the user when it is stale. No kirocrew-mate
tool writes the briefing, so none is named.
@rubencu
rubencu force-pushed the fix/briefing-item-no-file-tools branch from a8c833c to bef4d30 Compare October 20, 2026 14:25
@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author

Rebased onto main at 958316e and pushed bef4d30: #22128 conflicted with this PR in docs/architecture/context-management.md, in the capability table's briefing row. That row now names #22128's members_parts/prompt_layers.py and keeps this PR's clause about a member whose spec mounts no file tool. The code and test hunks are byte-identical to a8c833c.

That supersedes my PR Hygiene re-run request above: the new head's runs are waiting for approval instead.

@github-actions github-actions Bot added readiness: checking Automated validation is still running and removed merge conflict Branch has merge conflicts with its base — author must resolve before merge readiness: action required A blocking check or review needs attention labels Oct 20, 2026
@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author
  • Conductor crewmates lose nothing they need when a fresh session starts, with the DM thread as their only working memory

rebutted: checked. The DM thread is not a conductor's only working memory, and this PR does not change what carries over.

  • A conductor keeps its state in the ledgers, not in briefing.md. Every conductor spec auto-approves work_ledger_read / work_ledger_record (_CONDUCTOR_BASE_WORK_GRANTS) and session_ledger_read / session_ledger_record (_CONDUCTOR_CORE_GRANTS, _PIPELINE_CONDUCTOR_CORE_GRANTS).
  • The kirocrew-conductor charter, which the reported crewmates run, records its goal and next in the session ledger, drops its own chat at each round close with reset_conversation, and says "The next wake resumes from the ledger snapshot it carries" (_CONDUCTOR_SYSTEM_PROMPT in agent.py). That snapshot is prepended to every patrol cycle and re-injected on the first turn after a compaction (docs/architecture/context-management.md, "The [work ledger] snapshot").
  • A briefing already on disk is still injected on every fresh session, now under a header that names the file and says it may be out of date.
  • What the PR removes is only the instruction to keep briefing.md "with your file tools". A conductor could follow it only through execute_bash, which is outside its charter and an approval prompt on a default install.
  • Item 6 names the DM thread, as the Windows variant already does, because every template has one. Naming the ledgers for the templates that mount them would be a separate wording change.

@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author
  • On harnesses that ignore a spec's tools, conductor crewmates still have file writers but are told to drop briefing upkeep

rebutted: intended. The comment on _MEMBER_BRIEFING_ITEM_NO_FILE_TOOLS records it: the text speaks of the template, not of what the session can reach, because a harness that does not enforce tools still offers its own writers, and the template's no-write charter holds there too.

  • The charter says so on every harness: "A file to write, a build to run, a fix to make — each one is a work item for a child session. You have no file-writing tool" (_CONDUCTOR_SYSTEM_PROMPT), and the goal-conductor skill: "If a task needs a file written, it is a work item, not something you do."
  • So on such a harness, telling a conductor to keep its briefing "with your file tools" would ask it to do what its charter forbids, with a writer that only an unenforcing harness leaves within reach.
  • Choosing the wording by harness would make one template's charter hold on one harness and not on another. binding_has_file_write reads the template's own tools and excludedTools, so the answer is the same on every harness.

@rubencu

rubencu commented Oct 20, 2026

Copy link
Copy Markdown
Contributor Author

One red on bef4d30 is runner infrastructure, not this diff: "Decide whether the macOS suite runs" hit its 5-minute job timeout while actions/checkout was still fetching, so it never reached its path filter (run 39439060710, job 118687370768: "The job has exceeded the maximum execution time of 5m0s"). On a8c833c the same job passed and skipped the suite.

Could a maintainer re-run the failed job (gh run rerun 39439060710 --failed)? A fork contributor can't, and a push would put every run back behind approval.

@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 20, 2026
@KiroCrewCaptain
KiroCrewCaptain enabled auto-merge (squash) October 20, 2026 17:05

@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.

Tier 1 auto-approve: fix (4 files). Criteria: no conflict, no requested changes, security path denylist clean, design-doc gate clean, security checklist all-NO, no reviewer-lane findings, readiness signal fresh for this head SHA, declared category matches the diff. Category: a crewmate whose template mounts no file-writing tool is no longer told to maintain its briefing file; the new spec_has_file_write predicate picks the wording, with tests for both renderings. Spec files changed as a ride-along (a minority of the diff on both file count and changed lines), not reviewed as a design decision: docs/architecture/context-management.md.

@KiroCrewCaptain
KiroCrewCaptain merged commit 187aee5 into kirodotdev:main Oct 20, 2026
99 of 100 checks passed
@github-actions github-actions Bot removed the readiness: passed Eligible automated validation passed for the current revision label Oct 20, 2026
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.

2 participants