This repository was archived by the owner on May 21, 2026. It is now read-only.
[pull] main from awslabs:main#10
Merged
Merged
Conversation
…253) * docs(external-tool-integration): add Hermes Agent as worked example README.md already lists Hermes Agent alongside OpenClaw as a supported external harness for driving CAO via the cao-session-management skill, but external-tool-integration.md only walked through OpenClaw. Add Hermes target paths to Options A and C so the doc matches the README's supported-harness list. - Option A: add commented Hermes TARGET_SKILLS path (~/.hermes/skills/cli-agent-orchestrator) - Option C: add Hermes-specific skill_manage invocation and note the copy-vs-symlink staleness tradeoff - Intro: name both OpenClaw and Hermes Agent as worked examples Verified end-to-end on Hermes: launched code_supervisor + developer + reviewer via cao launch --headless --yolo on the kiro_cli provider and produced a working hello.py. * docs(skill): add prerequisites, profile discovery, and quick example to cao-session-management Agents loading the skill cold currently have to assemble a working 'cao launch' from rules scattered through the doc, and silently assume two prerequisites that fail confusingly when missing: - cao-server must be running at localhost:9889 - the target agent profile must already be installed via cao install Add three additive sections before 'Launching a Session': 1. **Prerequisites** — explicit cao-server health check (curl one-liner) and a reminder that profiles must be pre-installed. 2. **Discovering Available Profiles** — table of where to look depending on provider (kiro-cli agent list, agent-store/, README built-ins). 3. **Quick Example** — copy-pasteable end-to-end supervisor launch on the default kiro_cli provider, plus status/shutdown commands. No existing content removed; these are anchors a first-time agent can grab onto before reading the rules section. Verified against the skill end-to-end on Hermes Agent: launched code_supervisor + developer + reviewer headlessly and produced a working hello.py in 2m 38s. * docs(skill): make profile discovery and quick example provider-agnostic The previous skill update leaned too heavily on kiro_cli, but profiles in CAO are provider-agnostic — 'cao install' is the canonical install path for any provider, and ~/.aws/cli-agent-orchestrator/agent-store/ is the source of truth for what's installed locally. Workers inherit the launching terminal's provider unless pinned via frontmatter (README §Cross-Provider Orchestration). Changes: - Discovering Available Profiles: lead with the provider-agnostic agent-store/ listing and built-in install commands; demote 'kiro-cli agent list' to a provider-specific note (CAO mirrors profiles into ~/.kiro/agents/, which is why it works for kiro_cli). - Quick Example: keep kiro_cli as the default but enumerate every supported provider for --provider, and add a commented cross-provider variant so claude_code / codex / gemini_cli / kimi_cli / copilot_cli / opencode_cli / q_cli users see themselves in the doc. - Launching rule: replace the kiro-only --agents discovery hint with a cross-link to the new section. * Potential fix for pull request finding Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com> * docs(skill): correct profile discovery commands Per Copilot review on PR #253: `cao install <built-in>` writes to `agent-context/`, not `agent-store/`. The latter only holds custom local profile files. The HTTP endpoint `/agents/profiles` is the canonical, provider-agnostic way to see all installed profiles with `source` labels (built-in vs local). - Promote `curl /agents/profiles` as the recommended check - Split filesystem rows into agent-store (local) vs agent-context (built-ins) - Add explanatory note below the table * docs(skill): clarify Quick Example assumes a configured CAO setup Per @anilkmr-a2z review on PR #253: the Quick Example should foreground that it runs against an already-configured CAO host; `cao install` belongs to one-time setup, not to every launch. - Add note above the code block stating the example assumes a configured CAO setup and that users can skip to `cao launch` on an already-configured host. - Re-frame the install lines as Optional first-time setup rather than mandatory pre-launch steps. * docs(skill): refine /agents/profiles description for accuracy Per Copilot review on PR #253: the previous wording overstated the endpoint's behavior. `list_agent_profiles()` actually scans three sources (built-in packaged store, local agent-store, provider directories including agent-context) and de-duplicates by profile name (built-in wins), rather than "merging" two specific dirs. - Reword the table row to "all available profiles across built-in store + local store + provider directories" - Reword the explanatory paragraph to describe the three-source scan and dedup behavior accurately --------- Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.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 subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
See Commits and Changes for more details.
Created by
pull[bot] (v2.0.0-alpha.4)
Can you help keep this open source service alive? 💖 Please sponsor : )