docs(adr): graph-skill harness — Phase F F0 Wave 1#181
Open
ronsse wants to merge 2 commits into
Open
Conversation
Defines the substrate for running bounded "graph skills" against Trellis data. Five-tool surface (read_node, read_document, search_graph, propose_mutation, emit_event), markdown+frontmatter skill artifact, MutationExecutor as the only write path, five reserved SKILL_* event types. Mirrors adr-coding-agent-loop.md's structure; consumes the existing LLMClient Protocol so Anthropic + OpenAI work day one. Sibling to but distinct from src/trellis_workers/code_authoring/ — the tool surface is graph-internal, not filesystem/git. Substrate for F2 (curator), F3 (retrieval lazy enrichment), F4 (outcome feedback), F5 (score-based evolver). Phase F Wave 1 Unit A. Base SHA af374d2.
Disambiguates "graph skill" (the public category) from "skill" (in-doc shorthand) and contrasts the graph-skill harness in src/trellis_workers/agent/ with the coding-agent harness in src/trellis_workers/code_authoring/. Coordination edit on top of the harness ADR introduced in this PR. Co-Authored-By: Claude Opus 4.7 (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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Summary
docs/design/adr-graph-skill-harness.md(~264 lines) defining the substrate for Phase F's inner agent loop: bounded LLM-backed loops over Trellis data via a markdown-skill artifact + 5-tool allowlist (read_node,read_document,search_graph,propose_mutation,emit_event). Peer tosrc/trellis_workers/code_authoring/; distinct security model and tool surface.docs/design/adr-terminology.mddisambiguating "graph skill" (public category) from "skill" (in-doc shorthand) and contrasting the two harnesses.adr-coding-agent-loop.md. F1 implementation gated on this PR + sibling Wave 1 PRs merging.Test plan
adr-coding-agent-loop.mdadr-inner-curation-loop.md,workflow-engine-disposition.md)skill.dispatched,skill.step,skill.completed,skill.failed,skill.outcome_recorded) align with codebase dotted-lowercase convention🤖 Generated with Claude Code