agentsclimarketplace

Command routing

Skill event4u-app/agent-config/src/skills/command-routing

Use when the user invokes a slash command like /create-pr, /commit, /fix-ci, or pastes command file content — routes to the right command with context inference and GitHub API patterns.From its SKILL.md

Install
npx -y skills add event4u-app/agent-config --skill command-routing

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 19 stars19 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

5.4 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it

commands

When to use

Triggered when user invokes a slash command. The slash-command-routing-policy rule (always loaded) handles core behavior — this skill adds context inference and GitHub API patterns.

Procedure: Execute a command

  1. Match command — Find the command file in .augment/commands/ or agents/overrides/commands/.
  2. Infer inputs — Before asking the user, try to infer values (see table below).
  3. Execute steps — Follow the command steps in exact order.
  4. Verify output — Confirm expected result was produced (commit, PR, file change, etc.).

Before asking the user for input, try to infer it:

Input neededHow to infer
Jira ticketExtract from branch name (fix/DEV-1234-... → DEV-1234)
Default branchgit symbolic-ref refs/remotes/origin/HEAD or assume main
Project typeCheck for artisan (Laravel) or composer.json (Composer)
Module nameExtract from current working directory or file path
Current branchgit branch --show-current

Only ask the user if inference fails and the command cannot proceed without the value.

Command locations

LocationScope
.augment/commands/Shared commands (work across projects)
agents/overrides/commands/Project-specific overrides (used instead of original)

Commands that dispatch to the work engine

Most commands are pure markdown procedures — the agent reads the steps and executes them. Two commands delegate to the work_engine engine (TypeScript, run via tsx) via the ./agent-config dispatcher; both share the same Option-A loop (read state → run engine → handle exit code → repeat), they only differ in the input envelope they build:

CommandSubcommandEnvelopeUse when
/implement-ticket./agent-config implement-ticketinput.kind="ticket"User points at a Jira/Linear ticket or supplies a structured ticket payload
/work./agent-config workinput.kind="prompt"User supplies a free-form prompt — no ticket id, no acceptance criteria yet

Route prompt-shaped intents ("add a CSV export endpoint…", "fix the failing login test", "refactor the audit-log controller") to /work. Route ticket-shaped intents ("work on PROJ-123", "start on the ticket on this branch") to /implement-ticket. If the user pastes both a ticket id and a free-form goal, prefer /implement-ticket and let it pull the AC from the ticket — /work is the fallback when no ticket exists.

The actual step logic, halt formats, scoring breakdowns, and delivery report are emitted by the engine. Do not paraphrase or reorder engine output — surface it as-is. The two flows are mutually exclusive at the state-file level: one .work-state.json carries one envelope at a time, and the engine refuses to switch mid-flight.

The unified ./agent-config migrate command sweeps any legacy .implement-ticket-state.json file into the v1 .work-state.json schema as one of its cleanup steps (see docs/contracts/migrate-command.md). The wrapper invokes it automatically when the legacy file is detected; agents should not bypass the dispatcher.

GitHub API: Replying to PR review comments

When commands reply to PR review comments (e.g. /fix-pr-comments):

1. Read the setting

Read github.pr_reply_method from .agent-settings.yml:

ValueAPI call
replies_endpointPOST /repos/{owner}/{repo}/pulls/comments/{comment_id}/replies with {"body": "..."}
create_review_commentPOST /repos/{owner}/{repo}/pulls/{number}/comments with {"body": "...", "in_reply_to": comment_id}
autoTry replies_endpoint first. If it works → update setting. If 404/error → try create_review_comment.

2. Bot icon prefix

Read personal.pr_comment_bot_icon from .agent-settings.yml:

  • true → prefix reply body with 🤖 .
  • false or not set → no prefix.

3. API call rules

  • data must be clean JSON with ONLY required API fields — no summary or extra params.
  • Each reply is a separate API call — do NOT batch.
  • On first success with auto → update github.pr_reply_method in .agent-settings.yml to the method that worked.

Validate

  1. Verify all command steps were executed in order.
  2. Confirm expected output was produced (commit, PR, file change, etc.).
  3. Check that no step was skipped or improvised.

Output format

  1. Command executed — all steps completed in order
  2. Final result or summary as defined by the command

Gotcha

  • Don't ask questions when executing a command unless the command says "ask the user".
  • Don't add opinions during execution — follow steps exactly.
  • If a step fails, stop and report — don't improvise.

Do NOT

  • Do NOT ask questions during command execution unless the command says "ask the user".
  • Do NOT add opinions or summaries — execute steps exactly as written.
  • Do NOT improvise when a step fails — stop and report.

Auto-trigger keywords

  • slash command
  • command execution
  • agent command

What ships with it: 1 file

700 B alongside SKILL.md

evals/

Keep looking

Skills are one crate of 610,677. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.