docs: point the custom-* skills row at its activation on-ramp - #436
Merged
Conversation
The "Add your own skills" row in the INSTALL.md extension-points table told adopters where to put a custom skill and that it survives upgrades, but not that it stays inert until declared. Its sibling row (external knowledge base) already points at downstream-capabilities.yaml; this closes the one-cell asymmetry. Refs #292 Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Closes the documentation half of #292 — the split the issue's own triage analysis recommended shipping standalone as a
tiny-fix.docs/INSTALL.md's "Customizing without conflicts" table had a one-cell asymmetry between two sibling extension seams:knowledge_sources:in.agentcortex/context/private/downstream-capabilities.yaml.custom-*skill and that it survives upgrades — but not that survival is not activation.Verified against the code, an undeclared
custom-*skill is genuinely second-class:.agent/workflows/bootstrap.md:405— a skill id resolves only via the §3.6 rule table, the compact index, or acustom-*id declared indownstream-capabilities.yaml §skills; anything else is "unknown → ignore". So an undeclaredcustom-*id is never auto-recommended and cannot be pinned fromuser-preferences.yaml.docs/adr/ADR-007-...:33-38states the same outright ("Custom skills are second-class … cannot auto-activate").The row now says so and points at the existing opt-in on-ramp, mirroring the shape of its sibling row (pointer + ADR link + the
on-matchcap).Adopter delta: an adopter reading INSTALL now learns from the same table that a
custom-*skill needs a one-line declaration to become activatable. Engine behaviour is unchanged — no new detector, no auto-discovery (still rejected by ADR-007), no always-on cost.This deliberately does not build the runtime bootstrap detector the issue proposes; that half stays demand-gated per the issue's own report-trigger framing, so #292 is referenced, not closed.
Evidence
Validator —
bash .agentcortex/bin/validate.sh:Pinned test module (
docs/INSTALL.mdcarries substring assertions in this file, so it was re-run rather than assumed doc-safe):Table integrity re-checked after the edit (14 table rows before and after; single-line diff, no merged rows). Link target
docs/adr/ADR-007-downstream-capability-declaration-seam.mdconfirmed to exist.Scope: 1 file, 1 line changed, doc-only.
Refs #292
🤖 Generated with Claude Code