[6853] fix(frontend): Invoke registry skills by their name, not their slug

[ardaerzin]

Context

Typing / in the agent chat composer lists the agent's skills, but a registry skill showed up as /skill-name-23ad instead of /skill-name. The menu used the skill's registry slug, which carries a random suffix to keep workflows unique. The runner installs the skill under its own SKILL.md name, so the suffixed command never matched what the agent knows.

Changes

A new helper, skillCommandName, picks the command for each skill row. The embed entry already stores the skill's name next to the reference, so the helper uses that.

{"name": "legacy-inline-skill",
 "@ag.embed": {"@ag.references": {"workflow": {"slug": "legacy-inline-skill-18a1"}}, ...}}

Before: /legacy-inline-skill-18a1. After: /legacy-inline-skill.

If an entry has no valid name, the helper falls back to the slug, but only when the slug is itself shaped like a skill name. Built-in slugs (__ag__…, the older _agenta.…) never are. The only built-in entries saved without a name point at the retired getting-started stub in old revisions, so those rows are now hidden instead of showing a slug nobody should type.

No backend change. Registry slugs stay as they are.

Tests / notes

• 6 new unit tests in itemDescriptors.test.ts cover followed and pinned registry skills, the slug fallback, nameless and named built-ins, and inline skills.
• Checked live on /m against the EE dev stack: an agent with legacy-inline-skill (slug legacy-inline-skill-18a1) showed and inserted the suffixed slug before the change and /legacy-inline-skill after. Its other four skills were unchanged.
• The inserted text is still a hint for the agent, not a dispatch. That behavior is unchanged.

What to QA

• Add a registry skill to an agent, open a session, type /. The Skills section shows /<skill name> with no suffix, and Enter inserts that same text.
• Pin a skill to a version and repeat. Same result.
• Regression: an agent with an inline skill still lists it by its name.

[coderabbitai]

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository YAML (base), Organization UI (inherited)

Review profile: CHILL

Plan: Advanced

Run ID: 6d4bec1e-552a-4758-8878-1ec03564a8c5

📥 Commits

Reviewing files that changed from the base of the PR and between 28f3ca7 and 6385e00.

📒 Files selected for processing (5)

• web/packages/agenta-chat/src/hooks/useChatSlashCommands.tsx
• web/packages/agenta-entity-ui/src/DrillInView/SchemaControls/agentTemplate/itemDescriptors.tsx
• web/packages/agenta-entity-ui/src/DrillInView/SchemaControls/index.ts
• web/packages/agenta-entity-ui/src/DrillInView/index.ts
• web/packages/agenta-entity-ui/tests/unit/itemDescriptors.test.ts

🚧 Files skipped from review as they are similar to previous changes (1)

• web/packages/agenta-entity-ui/src/DrillInView/SchemaControls/index.ts

Included review availability: Your plan provides up to 8 included reviews per hour; 7 remain after this review.

---

📝 Summary

Summary by CodeRabbit

• Bug Fixes

  • Skill commands in the / palette now insert the correct invocation name, including names defined in skill metadata.
  • Improved handling for registry, built-in, pinned, and inline skills.
  • Invalid or missing skill names now fall back appropriately or remain unavailable instead of producing unusable commands.

• Tests

  • Added coverage for skill-name resolution across supported skill types and fallback scenarios.

Walkthrough

The change adds and exports skillCommandName, validates skill invocation names, tests resolution cases, and uses the helper for skill tokens in the chat slash-command palette.

Changes

Skill command name resolution

Layer / File(s)	Summary
Define and export skill command names web/packages/agenta-entity-ui/src/DrillInView/SchemaControls/agentTemplate/itemDescriptors.tsx, web/packages/agenta-entity-ui/src/DrillInView/SchemaControls/index.ts, web/packages/agenta-entity-ui/src/DrillInView/index.ts, web/packages/agenta-entity-ui/tests/unit/itemDescriptors.test.ts	skillCommandName selects a valid skill name and falls back to a valid embed slug. Exports and unit tests cover registry, pinned, built-in, and inline skills.
Use command names in the chat palette web/packages/agenta-chat/src/hooks/useChatSlashCommands.tsx	Skill rows now use skillCommandName(skill) instead of the removed skillToken helper. The comment describes inserted names rather than slugs.

Priority: ⬇️ Low

Estimated code review effort: 2 (Simple) | ~10 minutes

Merge Risk: ⚪ Minimal · up to 6385e

Skills without a valid command name are hidden from the slash-command palette, so no actionable merge risk remains.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	Docstring coverage is 100.00% which is sufficient. The required threshold is 60.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 4 functions across 5 files.
Linked Issues check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check	✅ Passed	Check skipped because no linked issues were found for this pull request.
Description check	✅ Passed	The description clearly explains the frontend change, the skill name resolution, fallback behavior, tests, and QA steps.
Title check	✅ Passed	The title clearly identifies the main change: registry skills are invoked by name instead of their suffixed slug.

✨ Finishing Touches

📝 Generate docstrings

• Create stacked PR
• Commit on current branch

🧪 Generate unit tests (beta)

• Commit to this branch
• Create a new PR

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}

[github-actions]

Railway Preview Environment

Status	Destroyed (PR closed)

Updated at 2026-09-17T13:37:03.852Z