Dogfooding `plan` + `implement` against a throwaway config surfaced one root
defect with many arms: the command templates treat consecutive fenced blocks as
one shell. They are not. Every ```bash fence runs as its own Bash call in its own
process, so a variable set in one block is empty in the next, and `$$` is a
different PID (measured: 21710 vs 22109).
The planner agent confirmed the sharpest arm at runtime, reporting that
`Mode: $RAW_FLAG` "arrived literally unsubstituted" — `--raw` was documented in
three command files while being functionally dead. A machine sweep found the same
root in 20 places across 9 files, well past the two the written fasit predicted:
- `$RAW_FLAG` read from non-shell agent prompts (analyze, plan, implement)
- `$TMPFILE` read across blocks (tokens, manifest, whats-active,
plugin-health) — each command could not read the file it had just written
- `$GLOBAL_FLAG` across blocks (fix)
- `$TODAY` never assigned in any block (campaign), passing
`--reference-date ""` to a write CLI in six places
- three `$$` temp paths handed to the Read tool (fix), which expands neither
All now follow the hardened drift.md pattern: a fixed literal path, or a
re-derivation inside each block that needs it.
Also fixed, all confirmed against ground truth rather than inferred:
- `implement` printed a rollback ID it never captured (the timestamp lived only
inside a command substitution) — the one message a user reads after a bad run
- `plan` reported "No analysis results found" for valid sessions, because Read
was pointed at a glob it cannot expand; now uses Glob and verifies the
analysis report exists before spawning the agent
- five phase commands wrote state.yaml with two of four required fields; since
the agent writes all four, a follow-up write silently deleted the rest
- `implement` promised rollback deletes created files; rollback deliberately
leaves them (M-BUG-26 still open) — the doc, not the engine, was wrong
- `implement` claimed a score delta with no pre-change measurement
- `verifier-agent` was told to write a report it has no tool to write
- dead `Task` tool name in always-loaded rule context; planner-agent template
demonstrated the inline file content its own line 110 forbids
The sweeps land as tests/commands/command-shell-state-shape.test.mjs, verified
red before the fix and proven able to fail by reintroducing the defect. Two
existing tests asserted the old bash-block mechanism rather than the intent and
were updated. Suite 1449/0; frozen v5.0.0 snapshots and all scanner code
untouched.
Not fixed, deliberately: neither command scope-gates its actions to the audit
target. The generated plan included an edit to a real file under ~/.claude,
outside the throwaway target, because the skill/agent scanners are machine-wide.
That is a design change, not a side fix.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_0195udHgCcFegzm7ecKku2Yc
Wave 5 Step 15. Threads --raw plumbing through all seven action
command templates and adds a shape test covering structural plumbing
plus help.md's plain-language vocabulary.
- commands/fix.md: --raw flag parsed; fix-plan rendering groups by
userActionLanguage; humanized title/description/recommendation are
rendered verbatim from the cross-referenced scan envelope.
- commands/rollback.md: terminology pass — "manifest" → "list of
changes" in user-facing copy; the file name manifest.yaml is kept
as the machine contract; --raw threaded through.
- commands/plan.md: --raw forwarded to the planner-agent's prompt;
agent now instructed to group actions by userImpactCategory and
lead with userActionLanguage; bash block added for flag parsing.
- commands/implement.md: --raw forwarded to the implementer-agent's
prompt; progress-log lines now reference the humanized titles
already present in the action plan.
- commands/cleanup.md: --raw accepted as no-op (cleanup is
file-management only, no findings prose); bash block added.
- commands/help.md: full plain-language pass — "PreToolUse" and
"frontmatter" jargon removed from user-facing copy; new
vocabulary table surfaces the humanized userImpactCategory and
userActionLanguage labels ("Configuration mistake", "Conflict",
"Wasted tokens", "Missed opportunity", "Dead config" / "Fix this
now", "Fix soon", "Fix when convenient", "Optional cleanup",
"FYI"); --raw documented as global pass-through flag.
- commands/interview.md: --raw accepted as no-op; "unused hooks"
question phrased as "unused automation that runs at specific
events" in user-facing copy.
tests/commands/action-commands-shape.test.mjs (new, +6 tests, 780 → 786):
- structural: bash block + Read tool + --raw/$ARGUMENTS plumbing
across all 7 files
- help.md vocabulary: ≥3 userImpactCategory labels and ≥3
userActionLanguage phrases present
- help.md jargon: no bare "PreToolUse" or "frontmatter" in copy
Add /ultraresearch-local for structured research combining local codebase
analysis with external knowledge via parallel agent swarms. Produces research
briefs with triangulation, confidence ratings, and source quality assessment.
New command: /ultraresearch-local with modes --quick, --local, --external, --fg.
New agents: research-orchestrator (opus), docs-researcher, community-researcher,
security-researcher, contrarian-researcher, gemini-bridge (all sonnet).
New template: research-brief-template.md.
Integration: --research flag in /ultraplan-local accepts pre-built research
briefs (up to 3), enriches the interview and exploration phases. Planning
orchestrator cross-references brief findings during synthesis.
Design principle: Context Engineering — right information to right agent at
right time. Research briefs are structured artifacts in the pipeline:
ultraresearch → brief → ultraplan --research → plan → ultraexecute.
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
