llp/0177-picker-enables-openclaw-without-attaching.issue.md is Status: Accepted on master with Extended-by: LLP 0180 (the fix decision), and its fix shipped in #594 (merged 4c44c8a). But no @ref LLP 0177 exists anywhere in the tree, so neutral observe reports it as backlog:
gaps: [{ "family": "pipeline", "target": "llp#177",
"reason": "uncovered request - needs a design" }]
backlog: [{ number: 177, ..., refs: [] }]
Why this is a live hazard, not bookkeeping
An uncovered request is the Designer's input. Acting on this gap would mint a fresh design LLP for "The init picker enables OpenClaw but never attaches it" and push integration/picker-enables-openclaw-without-attaching - designing a fix for a bug that shipped hours ago. The reconciler tick that filed this deliberately skipped it.
This is the same shape as #589 (LLP 0174 left at Accepted after its change set shipped, so it re-derived as implementable). Both are cases where the work is done but the machine-readable marker that makes "done" derivable was never written.
Why the existing references do not count
LLP 0180 cites 0177 heavily - **Related:** LLP 0177 (the issue this resolves), and LLP 0177 is resolved in its Consequences. But coverage is derived from @ref LLP NNNN annotations, not from Related: lines or prose, so none of it registers.
The fix
Annotate the code that resolves it. derivePickedClients (src/core/cli/walkthrough.js:1252) already carries @ref LLP 0180#decision [implements]; it wants a companion ref naming the issue that motivated the derivation. That is a ref which earns its place per CLAUDE.md (it tells a reader why the derivation exists at all, which neither the code nor the filename does), rather than a mechanical annotation added to satisfy a predicate.
Verification: neutral observe --json reports backlog: [] and no llp#177 gap once this lands.
Note for whoever picks this up
Until it merges, do not act on the llp#177 backlog gap and do not create integration/picker-enables-openclaw-without-attaching.
llp/0177-picker-enables-openclaw-without-attaching.issue.mdisStatus: Acceptedon master withExtended-by: LLP 0180 (the fix decision), and its fix shipped in #594 (merged4c44c8a). But no@ref LLP 0177exists anywhere in the tree, soneutral observereports it as backlog:Why this is a live hazard, not bookkeeping
An uncovered request is the Designer's input. Acting on this gap would mint a fresh design LLP for "The init picker enables OpenClaw but never attaches it" and push
integration/picker-enables-openclaw-without-attaching- designing a fix for a bug that shipped hours ago. The reconciler tick that filed this deliberately skipped it.This is the same shape as #589 (LLP 0174 left at
Acceptedafter its change set shipped, so it re-derived asimplementable). Both are cases where the work is done but the machine-readable marker that makes "done" derivable was never written.Why the existing references do not count
LLP 0180 cites 0177 heavily -
**Related:** LLP 0177 (the issue this resolves), andLLP 0177 is resolvedin its Consequences. But coverage is derived from@ref LLP NNNNannotations, not fromRelated:lines or prose, so none of it registers.The fix
Annotate the code that resolves it.
derivePickedClients(src/core/cli/walkthrough.js:1252) already carries@ref LLP 0180#decision [implements]; it wants a companion ref naming the issue that motivated the derivation. That is a ref which earns its place perCLAUDE.md(it tells a reader why the derivation exists at all, which neither the code nor the filename does), rather than a mechanical annotation added to satisfy a predicate.Verification:
neutral observe --jsonreportsbacklog: []and nollp#177gap once this lands.Note for whoever picks this up
Until it merges, do not act on the
llp#177backlog gap and do not createintegration/picker-enables-openclaw-without-attaching.