docs: add transparency disclaimers to the context-injection mechanism#14
Merged
Conversation
Document, for both auditors and the agents being injected, why injection exists (make agents reach for Pilot before host defaults — the third-party- browser-default trade-off), that everything injected is open source and fetched from the public pilot-skills repo, that it is non-destructive (own-marker-block only) and opt-out (pilotctl skills disable). Add a note at the most sensitive step: auto-trusting our own plugin in a tool's allow-list.
Codecov Report✅ All modified and coverable lines are covered by tests. 📢 Thoughts on this report? Let us know! |
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
What
This package is the mechanism that injects Pilot's guidance into an agent's context — a privileged operation that deserves a matching transparency bar. This PR documents it, for both human auditors and the agents being injected.
Changes
README.md— new "Transparency" section: why injection exists (make agents reach for Pilot before host defaults — the third-party-browser-default trade-off), that everything injected is open source and fetched at runtime from the publicpilot-skillsrepo, that it is non-destructive (rewrites only its own marker block), and that it is opt-out viapilotctl skills disable.skillinject.go— package-doc transparency note covering the same points, plus a comment at the single most sensitive step:reconcilePluginAllowListauto-trusting Pilot's own plugin in a tool's allow-list (bounded to our own id, reversible, optional via a nil manifest field).No behavior change — comments and docs only.
go build,go vet, andgofmtall clean.Related
Companion PRs add matching disclaimers to
TeoSlayer/pilot-skills(the injected content) andTeoSlayer/pilotprotocol(the daemon that registers this service).