Merged
Conversation
…show metadata-only in system prompt https://claude.ai/code/session_01P24JxNnhgVDgwupZ9J1M31
…elative_path - Canonicalize resolved target path to prevent symlink escapes out of skills dir - Rename 'path' parameter to 'relative_path' for consistency with write_skill_file - Add subdirectory example to parameter description https://claude.ai/code/session_01P24JxNnhgVDgwupZ9J1M31
- Updated CLAUDE.md to clarify the distinction between instruction and subagent skills. - Added detailed description of the daily-news-to-threads orchestration skill and its functionality. - Revised README.md to reflect the new orchestration and subagent capabilities in bot skills. - Enhanced logging in agent.rs for better debugging of subagent tool usage and iterations.
…uction and subagent handling - Enhanced logging in LlmClient to include model information in debug output. - Revised build_context method to provide clearer metadata for instruction and subagent skills, emphasizing the use of read_skill_file for instruction skills. - Updated tests to reflect changes in context generation, ensuring instruction skills do not inject full content but provide hints instead.
…ontext Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…nd header assertions Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…g tools at launch Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
…n-efficiency rules
…tion, standardize archetype terminology https://claude.ai/code/session_01Hxm7xsx3RZ93yUgWPQbVsK
…val escape hatch, add soul.v to field guide
…5OTEa' into claude/agent-skills-personality-OofJq
…subagent SOUL.md: rewrite from verbose YAML to compact key=value notation - Three tiers: CORE (immutable), STYLE (semi-mutable), CTX (mutable) - key=value with | separator between fields, , separator within values - ~40 tokens vs previous ~120 tokens — 3x reduction - Fix description to third-person trigger condition (not behavioral instruction) - Add Tensions section for nuanced character expression soul-keeper/SKILL.md: convert from instruction skill to subagent skill - Add model/tools/max_iterations frontmatter → only metadata in system prompt - ~20 tokens per conversation vs ~200 tokens for full body injection (10x reduction) - Add canonical SOUL.md template for model alignment - Add before/after example showing minimum-change discipline - Add Field Guide table mapping signals to fields and tiers https://claude.ai/code/session_01Hxm7xsx3RZ93yUgWPQbVsK
Four violations introduced by the PR #21 merge were not auto-formatted: - agent.rs:155 blank line / trailing whitespace mismatch - agent.rs:704 info!() macro args collapsed to one line - agent.rs:1373 vec![] literal broken to multi-line (line-width) - llm.rs:106 debug!() call broken to multi-line (line-width) cargo fmt --all -- --check now exits 0. https://claude.ai/code/session_01Hxm7xsx3RZ93yUgWPQbVsK
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.
No description provided.