The model was offering "commit to current branch" on main instead
of "create a branch and PR." Added explicit branch detection step
and "Do NOT commit directly to main" instruction.
Let the agent generate a reasonable branch name based on context
and repo conventions instead of prescribing a date-based format
that would collide on multiple runs per day.
Handles committing changes at the end of a refresh run so doc
maintenance doesn't sit uncommitted. Detects git context and adapts:
autonomous mode uses sensible defaults (branch + PR on main, separate
commit on feature branches), interactive mode presents options. Always
selectively stages only compound-refresh files to avoid mixing with
in-progress feature work.
The file-tools-over-bash instruction was in the orchestrator's
context but not passed to spawned subagents. Changed to an explicit
quoted instruction block that must be included in each subagent's
task prompt so it's visible to the subagent, not just the orchestrator.
Auto-archive now requires both the implementation AND the problem
domain to be gone. If referenced files are deleted but the application
still deals with the same problem (auth, payments, migrations), the
learning should be Replace'd not Archive'd — the knowledge gap needs
to be filled. Uses agent reasoning about concepts, not mechanical
keyword searches.
The model was generating findings internally then outputting a
one-line summary. Added explicit instructions that the full report
must be printed as text output — every file, every classification,
every action. In autonomous mode, the report is the sole deliverable
and must be self-contained and complete.
- Restructure Phase 3 with explicit autonomous skip section that says
"do not ask, do not present, do not wait" before any interactive
instructions
- Add autonomous caveats to Core Rules 4, 7, 8 which previously had
unconditional "ask the user" language
- Clarify that missing referenced files is unambiguous Archive evidence,
not a doubt case requiring user input
Instead of requiring write permissions, autonomous mode attempts
writes and gracefully falls back to recommendations when denied.
Report splits into Applied (succeeded) and Recommended (could not
write) sections. Read-only invocations produce a maintenance plan.
Support mode:autonomous argument for unattended/scheduled runs.
In autonomous mode: skip all user questions, apply safe actions
directly, mark ambiguous cases as stale with conservative confidence,
and generate a detailed report for after-the-fact human review.
- Broad scope triage: inventory + impact clustering + spot-check drift
for 9+ docs, recommends highest-impact area instead of blind ask
- Drift classification: sharp boundary between Update (fix references
in-skill) and Replace (subagent writes successor learning)
- Replacement subagents: sequential subagents write new learnings using
ce:compound's document format with investigation evidence already
gathered, avoiding redundant research
- Stale fallback: when evidence is insufficient for a confident
replacement, mark as stale and recommend ce:compound later
- Use platform-agnostic interactive question tool phrasing with examples
for Claude Code and Codex instead of hardcoding AskUserQuestion
- Fix contradiction between Phase 2 auto-archive criteria and Phase 3
always-ask-before-archive rule so unambiguous archives proceed without
unnecessary user prompts
Adds a new skill that reviews existing docs/solutions/ learnings against the
current codebase and decides whether to keep, update, replace, or archive them.
Also enhances ce:compound with Phase 2.5 selective refresh checks.
Co-Authored-By: Claude <noreply@anthropic.com>
Name specific blocking question tools (AskUserQuestion, request_user_input,
ask_user) so agents actually invoke them instead of printing questions as
text output. Updates skill compliance checklist to match.
Add a highest-leverage-move question to the product pressure test,
a challenger option in approach exploration, and a low-cost change
check to the finalization checklist.
Rebased onto main after #251 restructured commands into skills/.
Applied the same namespace fix to the new skill file locations
(ce-brainstorm, ce-plan, ce-review SKILL.md files).
Short agent names like `repo-research-analyst` are replaced with
fully-qualified `compound-engineering:research:repo-research-analyst`
to ensure correct agent resolution across all plugin targets.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Update CLAUDE.md, AGENTS.md, plugin CLAUDE.md, and versioning docs
to make clear that contributors should not hand-bump plugin versions
or cut changelog entries in normal PRs. The automated release process
decides the next version after choosing which merged changes ship
together.
Anthropic now maintains a better skill-creator at
github.com/anthropics/skills/tree/main/skills/skill-creator.
Removes the bundled skill-creator and updates component counts
from 47 to 46 skills.
The /technical_review command was removed in v2.32.0 but references
remained in ce:plan and deepen-plan menus. Remove them entirely -
the existing 'Review and refine' option (document-review) already
covers plan review.
Fixes#244
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Move all 27 command .md files from commands/ to skills/*/SKILL.md
format. Claude Code 2.1.3+ merged commands and skills - both create
slash commands and work identically. Skills add optional features
like supporting files and automatic context loading.
- commands/ce/*.md -> skills/ce-*/SKILL.md
- commands/workflows/*.md -> skills/workflows-*/SKILL.md
- commands/*.md -> skills/*/SKILL.md
- Update plugin.json and marketplace.json descriptions
- Update CLAUDE.md directory structure docs
- Update deploy-docs and generate_command references
Fixes#116
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The `create-agent-skills` SKILL.md contained literal `!`command`` dynamic
context injection directives as documentation examples. Claude Code's
preprocessor executes these directives as plain text before the skill
content is sent to the model — it does not parse markdown, so fenced code
blocks and inline code spans offer no protection.
When loading this skill, the preprocessor attempted to run `command`,
`gh pr diff`, and `gh pr diff --name-only` as shell commands, producing:
Bash command failed for pattern "!`command`":
(eval):1: redirection with no command
This caused the skill to fail silently or error on every invocation.
Fix: replace the literal directives with a prose description of the
syntax, add a warning about the preprocessor behavior, and link to
`references/official-spec.md` § "Dynamic Context Injection" for the
concrete example (reference files are loaded on-demand by Claude via the
Read tool and are not preprocessed).
Upstream context:
- anthropics/claude-code#27149 (closed, wontfix): preprocessor correctly
executes bare text; workaround is to describe the syntax in prose
- anthropics/claude-code#28024 (closed as duplicate of #13655)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Bump plugin to 2.39.0 with features from Matt Van Horn (@mvanhorn):
context budget precheck, plan sequence numbers, review serial mode,
agent-browser debugging commands, test-browser port detection, lfg
phase gating, and Context7 API key auth.
Also fixes MCP server merge order so plugin servers correctly
overwrite stale entries during sync.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Adds --serial flag and auto-detection (6+ agents) to run review agents
sequentially instead of in parallel, preventing context limit errors
with Opus 4.6.
Closes#166
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Documents all missing agent-browser CLI commands for debugging workflows.
Adds 9 new command categories: eval, console/errors, network, storage,
device settings, element debugging, recording/tracing, tabs, and
advanced mouse controls.
Closes#170
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Add Phase 0 context budget check that warns users when running /compound
near context limits, and offers a compact-safe single-pass alternative
that avoids launching 5 parallel subagents.
Closes#198
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
The every-style-editor agent file was a duplicate of the existing
every-style-editor skill. Agent files are eagerly loaded into the
Task tool definition on every API call (~100-200 tokens each),
while skills are lazy-loaded only when invoked. Removing the
duplicate saves tokens and eliminates potential runtime errors
when the agent is invoked via Task tool instead of Skill tool.
Changes:
- Delete agents/workflow/every-style-editor.md (skill version
in skills/every-style-editor/ already exists)
- Update README.md workflow agent count from 5 to 4
- Update plugin.json agent counts from 29 to 28
Fixes#156
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Replace all hardcoded localhost:3000 references with dynamic port
detection. The command now checks (in priority order): explicit
--port argument, CLAUDE.md config, package.json scripts, .env files,
then falls back to 3000.
Closes#164
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
